BSG CSSC Service Development Prize 2023 highly commended submission by Prof David Sanders and Dr Jeremy Woodward

What were the challenges to your service and why did you need to change?

Coeliac Disease affects 1% of the adult population but currently about 10,000 cases are diagnosed per annum in the UK. For most patients their symptoms will improve on a gluten-free diet (GFD). However, a percentage may have non-responsive symptoms and will require further investigations. Refractory coeliac disease (RCD) is rare and is defined as persistent or recurrent malabsorptive symptoms/ signs with villous atrophy, despite adherence to a strict GFD for at least 12 months. However, RCD may be suspected before this as patients with RCD often show little response to a GFD, meaning these patients may present with severe and/or progressive malabsorptive symptoms before this 12-month GFD trial period.

RCD has a reported prevalence of 0.3-4% of patients with coeliac disease and non-responsive coeliac disease (NRCD) is reported to account for 8-23% of cases. This wide range may reflect the difficulty in differentiating between inadvertent gluten exposure, slow responders, individuals with super-sensitivity and true RCD, meaning RCD may be over diagnosed the true prevalence of RCD is likely to be lower.

Individuals with RCD can be sub-classified as having either RCD Type I (RCD1) or Type II (RCD II), based on the abnormal expansion of a subset of small intestinal intraepithelial lymphocytes (IELs), which is detected in RCD II. RCD II is predominantly diagnosed in adults aged 50 or above, although younger cases have been observed.

Given that the prevalence is so low Nationally GI Consultants have very limited exposure to RCD I or II. From a clinicians’ perspective this means that you may only see a few cases of NRCD or RCD throughout your Consultant career and perhaps only 1 case of ‘true’ RCD II. From a patients’ perspective once they have been told that they have RCD internet searches that they undertake reveal that the mortality is greater than 50% at 5 years and they view this as a ‘death sentence’. Understandably, they are unable to discriminate between RCD I and RCD II and this is not surprising given how difficult this is for us to do as practising GI clinicians!

This means that there is an absence of centralised expertise in terms of managing such patients in the UK and this is a challenge to the UK NHS GI service. Furthermore there is also an unmet need from the patients perspective (Coeliac UK James Lind Alliance patients priorities number 2 and 10) www.jla.nihr.ac.uk/priority-setting-partnerships/coeliac-disease/top-10-priorities.htm

How did you overcome the challenges? 

Sheffield GI Consultants have worked collaboratively with the specialist GI dietetic team (Nick Trott, Rachel Buckle and Christian Shaw) and other member of staff (secretaries, clerical and medical team) to provide a bespoke national service for adult patients with NRCD and RCD.

You can access this service at different levels.

1. We provide an immediate contact telephone or email service for any clinicians who are just seeking advice and wish to speak to a Consultant with experience in this disease (>3000 adult patients with coeliac disease over 20 years).
2. When considering RCD I we have introduced urinary gluten immunogenic peptides (GIP) into UK practice. GIP’s detect the consumption of gluten which is then excreted in the urine. Patients with established RCD1 and persisting mucosal inflammation and/or ongoing symptoms provided three urine samples for GIP analysis.
3. We also provide a direct referral service to rapidly assess such patients. In order, to differentiate between Type I and Type II RCD we have set up and pioneered a small bowel biopsy flow cytometry service (on normal saline samples). This service is only available in our 2 centres within the UK. Patients are then contacted personally and their management plan and admission is tailored according to their individual needs and geographical location. Patients are admitted with rapid access for their tests, whether this is capsule endoscopy, enteroscopy, clinical and dietetic assessment. Thereafter, rapid follow-up contact occurs with results and subsequent therapeutic interventions if necessary including steroids, immunosupressives and chemotherapy.

What were the outcomes? 

To date we have been referred 285 adult (≥16 years) patients with NRCD or RCD.

The commonest cause of NRCD across the cohort was gluten exposure (72/285; 25.3%). RCD accounted for 65/285 (22.8%) patients; 54/65 (83.1%) had RCD1 and 11/65 (16.9%) patients had RCD2. The estimated 5-year survival for RCD1 was 90% and 58% for RCD2 (p=0.016). 36/54 (66.7%) patients with RCD1 underwent urinary GIP testing. 17/36 (47.2%) RCD1 patients had at least one positive urinary GIP test.

Outcome 1: The majority of patients referred from across the UK do not have RCD II. This is a hugely important and reassuring finding for referring clinicians. From a patients’ perspective this is a huge relief for them to learn that they are unlikely to develop the serious complications or have a significantly reduced life expectancy.

Outcome 2: The high frequency of urinary GIP positivity suggests that gluten exposure may be common in RCD1; this may avoid the unnecessary use of steroids or immunosuppressants in this group of patients.

Outcome 3: We have reported the lowest international mortality for RCD II. Rapid access and novel treatments such as cladribine, campath, interleukin 15 and more recently stem cell transplant may play a role in this outcome.

Outcome 4: Referring Clinician Feedback > 95% said good or excellent on a Likert Scale 1-5. The commonest free text comment ‘patients should be referred to a recognized National Centre for consideration of novel therapies’.

Outcome 5: As a result of pioneering this service NHS England have recognised us as the National Centre: Rare Disease Collaborative Network (Sheffield as lead and Dr Jeremy Woodward, Cambridge as member).

Outcome 6: Our Unit has consistently had the highest scores in Sheffield Teaching Hospitals for the Friends and Family Test.

What were the learning points and how can this influence other teams? 

This is an entirely new UK service for what could be called a ‘rare disease’. The lessons learnt are applicable to any GI and Liver clinicians aspiring to provide a national service but these learning points could also be attributed to planning of regional services.

There were many crucial learning points:

1. We are truly a multi-disciplinary team and that approach has been crucial to provide the best service for the patients coming to us for a short intensive period from across the UK. It is a very proactive team who always ensure a warm and welcoming approach is guaranteed, to try and alleviate the understandable concerns and anxieties that such patients face. The team joke (but in many ways are entirely serious) that by adopting a John Lewis approach or viewing every patient as your relative, ensures the best care is achieved. We would like to emphasise that this service was built from scratch and we embraced the new addition to our workload.
2. It was crucial to have ‘buy in’ from clinicians across the UK. We did this through rapid response and access as well as advertisement (through medical meetings) and publications (review articles about the service with explicit contact details and how to refer sections: Baggus EMR et al Frontline Gastroenterology and Penny HA et al Nutrients 2020).
3. We published our outcomes which we think this gave confidence to clinicians and patients alike (Penny HA et al Nutrients 2022).
4. We directly engaged with the National Patient Charity and NHS England and sought their guidance and endorsement.

We would like to thank all the colleagues and clinicians who referred patients to us and allowed us to have this opportunity and gain this expertise.