Compositional analysis of exclusive enteral nutrition and its effects on the paediatric Crohn’s disease faecal microbiome and metabolome during treatment and food reintroduction

Logan, Michael (2020) Compositional analysis of exclusive enteral nutrition and its effects on the paediatric Crohn’s disease faecal microbiome and metabolome during treatment and food reintroduction. PhD thesis, University of Glasgow.

Due to Embargo and/or Third Party Copyright restrictions, this thesis is not available in this service.


Crohn’s disease (CD) and ulcerative colitis (UC) are the two major types of inflammatory bowel disease (IBD). Approximately 10% of patients with CD are diagnosed during adolescence or early adulthood. Several hypothesises have been proposed to explain the rising incidence in CD worldwide. In particular, the increasing use of food additives in food industrialisation has been the subject of intensive research. Results from these studies have quickly been translated to dietary recommendations and exclusions.
The importance of nutrition in patients with CD has been described in several aspects of the disease, from aetiology, management, and the long-term health outcomes of the patient. Nutritional therapy in the form of exclusive enteral nutrition (EEN), is the first line treatment for the induction of remission for active disease in paediatric CD patients throughout Europe, Australia, and parts of North America. Exclusive enteral nutrition is effective in inducing clinical remission and has the additional benefit of restoring nutritional deficiencies.

Exclusive enteral nutrition has been associated with greater levels of mucosal healing compared with corticosteroids, the next most common therapy for the induction of remission. Taking the extensive clinical evidence of EEN, it was hypothesised that food additives contained within these formulas were less likely to initiate disease flares, and that their dietary exclusion is not necessary for the amelioration of active disease. A study was performed to document each EEN formula with published clinical efficacy for the induction of remission in active CD. Subsequently, a nutrient and food additive compositional analysis was performed. This analysis revealed that all EEN formulas contain a high number of food additives, many of which have been previously implicated in the pathogenesis of IBD. In particular, modified starches, which include maltodextrin, were included in every EEN formula identified. Many of these food additives have been implicated in the pathogenesis of IBD through in vitro and in vivo experiments. Despite these experimental studies showing the potentially negative influence of food additives, the extensive clinical evidence of EEN formulas, all of which contain food additives, highlights the difficulty in replicating experimental evidence in the clinical setting. As many patients experience significant food avoidance and have poor food-related quality of life, this study may help alleviate some of this food aversion.

Faecal calprotectin (FC) is currently the leading non-invasive biomarker of intestinal inflammation and mucosal healing. Exclusive enteral nutrition has been shown to effectively reduce the levels of FC during the course of treatment. It has also been shown that levels of FC revert to pre-treatment levels within the first four months of return to habitual diet post EEN completion. However, it is unknown how rapid this increase in FC occurs; and the extent to which maintenance therapies may be able to delay the rise.

Despite its first use in the 1970’s, the precise mechanisms of action of EEN remain elusive. Recent evidence has implicated the role of the intestinal microbiome. Exclusive enteral nutrition has been shown to lead to the reduction in concentration of certain short chain fatty acids (SCFA), to which numerous studies have attributed anti-inflammatory properties. Exclusive enteral nutrition has also been shown to increase faecal total sulphide concentrations and decrease faecal pH from acidic levels to alkaline. These counterintuitive findings have also been demonstrated with reductions in specific bacterial species, including Faecalibacterium prausnitzii, a major butyrate producer which has been suggested as a protective species in CD. As with changes in FC, many of the microbial and metabolic changes observed during EEN have been shown to revert to pre-treatment levels within the first four months of food reintroduction.

It was the aim during this PhD to investigate how changes that occur in the faecal microbiome and metabolome during EEN are maintained as patients return to their habitual diet. During this study the researcher recruited a cohort of paediatric CD patients with active disease prior to starting a course of EEN. Up to five stool samples were collected from each of these patients. An initial stool sample was collected from patients prior to starting EEN; two further stool samples were collected during EEN (33d & 54d EEN); and finally, two stool samples were collected during food reintroduction, the first within two weeks, and the second within two months of food reintroduction. Chapter 4 of this thesis describes a study in which serial FC measurements were performed on these faecal samples. This study shows that FC is significantly reduced during EEN but within only 17d food reintroduction, FC levels have increased to almost baseline levels. The results from this study have also shown that patients consuming maintenance enteral nutrition (MEN) had a lower concentration of FC during the first 17d of food reintroduction than patients not using MEN. This benefit was not maintained in the longer-term phase of food reintroduction, and FC had reverted to pre-treatment levels within two months of food reintroduction.

The current study also describes changes in the faecal microbiome and metabolome from baseline, during EEN and subsequent food reintroduction. Comparisons between the faecal microbiome and metabolome of patients with CD were also made against a cohort of healthy children, patients with UC, and participants who underwent an exploratory colonoscopy for the diagnosis of IBD but after investigation did not have an IBD diagnosis. The faecal microbiome was assessed with molecular microbiology techniques through amplification of the 16s rRNA gene, and the faecal metabolome was characterised using several techniques including gas-chromatography and 1H nuclear magnetic resonance. Patients with CD had reduced species diversity and richness compared with healthy participants. During EEN this reduced diversity, relative to healthy controls, was maintained but had not changed from baseline levels. When β-diversity was assessed using Bray-Curtis dissimilarity matrix, the community structure was shown to be driven further from its position at baseline, and in the opposite direction from the HC. However, when this analysis was repeated using the unweighted UniFrac distance measure, while distinct from HC at baseline, the faecal microbiome moved to be more similar with HC by the end of EEN. During food reintroduction there was no change in β-diversity from the position at the end of EEN, when assessed using the unweighted or weighted UniFrac. However, a significant shift in β-diversity was observed during food reintroduction when β-diversity was assessed using Bray-Curtis dissimilarity.

During EEN and food reintroduction several taxa changed significantly in relative abundance. Several of the same bacterial taxa were highlighted for their importance when using advanced bioinformatic techniques including sparse projection to latent structure – discriminant analysis (sPLS-DA), random forest modelling and during correlation analysis with FC. More specifically, these analyses repeatedly implicated genera from Faecalibacterium, Haemophilus, and Veillonella as important in driving changes in the faecal microbiome community structure and correlating significantly with absolute concentrations of FC.

In summary, the results from this PhD challenge current perceptions surrounding the role of food additives and their role in the management of CD. It is hypothesised that the additives contained within EEN formulas are unlikely, at least in the concentrations provided in the formulas, to represent significant dietary triggers to CD. As many patients with CD experience significant food aversions and poor food quality of life, this list of food additives that are unlikely to act as triggers of CD may help alleviate some of this food aversion. This study has again confirmed the earlier evidence of the significant reduction in FC during EEN, which is paralleled with reduction in SCFAs, faecal pH and increased faecal total sulphide. As FC increases rapidly within 17d of food reintroduction, current disease maintenance strategies are inadequate at preventing the development of subclinical inflammation. Future research should explore dietary triggers of CD relapse during the food-reintroduction phase and their interplay with the gut microbiome and metabolome.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Crohn's disease, exclusive enteral nutrtion, immunology, inflammation, nutrition, paediatric, inflammatory bowel disease, microbiome, bioinformatics.
Subjects: Q Science > QR Microbiology
Q Science > QR Microbiology > QR180 Immunology
R Medicine > RJ Pediatrics
Colleges/Schools: College of Science and Engineering > School of Engineering
Supervisor's Name: Ijaz, Dr. Umer Zeeshan
Date of Award: 2020
Embargo Date: 2023
Depositing User: Michael Logan
Unique ID: glathesis:2020-81262
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 07 Apr 2020 08:49
Last Modified: 07 Apr 2020 14:01
Thesis DOI: 10.5525/gla.thesis.81262
Related URLs:

Actions (login required)

View Item View Item