MAgnetic Resonance Imaging in COeliac Disease

Hypotheses of the study

Based on the investigators' pilot data and on the literature available, this study aims to test the main hypotheses that in adults newly diagnosed with coeliac disease treatment with GFD will:

1. reduce the water content of the fasting small bowel
2. reduce the volume of the fasting colon
3. increase whole gut transit time

Objectives of the study

The specific objectives of the study are therefore to quantify any change following treatment with a gluten-free diet (GFD) in:

1. the water content of the fasting small bowel
2. fasting colon volume
3. whole gut transit time in adults newly diagnosed with coeliac disease using non-invasive MRI Co-variates of interest

During this study investigators will measure a series of exploratory co-variates of interest at diagnosis and at 12 months follow-up:

1. BMI
2. Gastrointestinal symptoms (Bristol Stool diary for a week, HAD, PHQ-15, IBSS, QOL, GI symptoms VAS scales )
3. Marsh grading of D2 segment duodenal biopsies (defined as routine clinical assessment in NUH)
4. Fasting breath hydrogen
5. Serology values (tTG, EMA)
6. Faecal microbiota and short chain fatty acids

In a subgroup analysis, investigators will describe associations (if any) between the MRI outcomes and these co-variates:

1. before treatment,
2. after treatment and
3. differences between the two.

The latter will represent the most useful analysis. By using the after treatment measure as the outcome and the before treatment measure as a covariate this will be mathematically equivalent to analysing the change with treatment (adjusted for baseline).

Parallel Healthy Volunteers group This study will also include a parallel pilot study healthy volunteers (HVs) frequency matched for age (in 20 years bands) and gender to provide descriptive statistics on a likely reference range for the healthy population to inform the design of future studies.

A pilot study in 36 newly diagnosed coeliac disease patients before and after a gluten-free diet.

University based

Pilot study on patients:

There is no data available to estimate the size of the change in water content of the fasting small bowel in patients hence this is a pilot study.

The investigators' own MRI pilot data in n=20 untreated, newly diagnosed coeliac patients showed a fasting small bowel water content of (mean ± SD) 202 ± 115 ml volume. We can predict that we should be able to detect a change of 40% (or a reduction of 80 ml volume) after gluten free diet with a power of 90% and a Type I error probability of 0.05 using n=24 patients in a paired study design. This is considered a clinically significant change after gluten free diet; this change reflects twice the level of normal control values of 65 ± 43 ml volume. Investigators plan to recruit n=36 to allow for dropouts and to increase power also for secondary outcomes.)

Pilot study on healthy volunteers (HVs):

For exploratory purposes and to provide pilot reference data for the design of future studies, investigators aim to collect data on 36 HVs 36 patients and 36 healthy volunteers.

Pilot study in coeliac disease:

Investigations include:

1. Magnetic Resonance Imaging - 2 scans 12 months apart
2. Self-administered symptom questionnaires
3. Stool sample
4. Hydrogen breath test The patients will have had a biopsy and serology as part of their standard care. The healthy volunteers will undergo only the blood sample as part of screening with no biopsy taken.

Duration of study = 36 months. Duration of study for each participant = around 12 months. Follow-up of participants will stop after 12 months if they fail to receive the subsequent MRI scan There is no randomisation or blinding as this is a simple pilot study. Primary endpoint

1. Fasting small bowel water content (SBWC) measured in ml approximately 12 months after initiation of GFD Secondary endpoints
2. Fasting colon volume measured in ml
3. Whole gut transit time measured in hours

Co-variates analysis as indicated above

The basic characteristics of the study population will be calculated using frequencies and proportions.

Objective 1:

For this cohort of patients with newly diagnosed coeliac disease, investigators will compare the mean difference of their fasting small bowel water content (the primary outcome) before and after the initiation of GFD. This will be done using Paired Test at 5% level of significance. This assumes normality of the data, should the data have a high level of skewness non-parametric methods will be used. In order to assess the potential effect modification with age and gender, investigators will stratify the above analysis by those variables. For this purpose age will be considered as categorical variable and analysed as 20 years bands. This analysis will only be conducted for cases with complete follow-up data on MRI scan.

Objectives 2 and 3:

This analysis will be repeated for the two secondary outcomes, namely the volume of the fasting colon and the gastrointestinal transit time.

Co-variate analyses in the coeliac disease group. The potential association between co-variates (gastrointestinal symptoms, Marsh grading etc.) and the three MRI outcomes will be assessed using stepwise linear regression analysis. Investigators will also assess the magnitude of the effect in term of mean difference along with their respective confidence interval. The analysis would be separately conducted for each outcome and co-variates recorded during the first and the subsequent MRI scan.

Should this analysis highlight a few particularly strong associations, investigators will attempt to input those to a regression model to predict either baseline values or change with GFD.

Pilot study in HVs The HVs group is included to provide pilot reference data to inform the design of future studies. Investigators will compare the small bowel water content among the coeliac disease cases to the HVs (controls), and then compare the mean difference pre- and post-measurement between cases and controls using linear regression.