Effectiveness of Exclusive Enteral Nutrition in Crohn's Disease

Abstract and Introduction

Abstract

Background: Exclusive enteral nutrition (EEN) induces remission and mucosal healing in patients with Crohn's disease, but the mechanism of action remains unknown.

Aim: To outline current understanding of the mechanisms of action of EEN.

Methods: From a comprehensive literature search, published data were critically examined in a narrative review.

Results: Multiple potential mechanisms of action have been identified. EEN optimises nutritional status. Differences in gut microbiota in terms of overall diversity and taxonomic community structure are observed between responders and non-responders to EEN. Therapy with EEN alters microbial metabolites (including faecal short-chain fatty acids, amino acids, branched-chain amino acids and sulphide) and faecal pH. Epithelial effects and restoration of barrier function, as well as changes in mucosal cytokine profiles and T-cell subsets are observed in responders to EEN. The impact of inclusion or exclusion of specific dietary components may be of importance, but putative detrimental components are found in many formulas. A major challenge in interpreting these findings is that they often contradict or change in opposite directions to what is considered 'beneficial'. It is difficult to differentiate between the observations following EEN being driven by EEN per se and those associated with resolving inflammation.

Conclusions: The mechanisms of action of EEN are likely to involve a complex interplay between host mucosal immune response and luminal environment, but the identity of key factors remains poorly understood. A better definition of pathogenic factors may aid in developing more targeted dietary treatment and provide insights into the pathogenesis of Crohn's disease.

Introduction

Therapeutic strategies for control of intestinal inflammation in patients with Crohn's disease (CD) are mostly pharmacological. The introduction of biological therapies, in particular, has changed the management of CD considerably and enabled a shift in therapeutic goals from control of symptomatic disease to control of inflammation itself, with endoscopic mucosal healing being a major target of disease treatment. Since inflammation in CD is complex with many pathways identified, there has been no shortage of candidate factors or methodologies to inhibit or promote the inflammatory process. However, despite such progress, therapies that directly target cells or factors involved in the inflammatory process are suppressive at best and lead to mucosal healing in a minority of patients.

Exclusive enteral nutrition (EEN) is best described as the provision of nutritionally complete liquid formula, without exposure to other foods, to patients with CD, usually for 6–8 weeks.^[1] Its success in inducing both clinical remission and intestinal mucosal healing in a high proportion of patients with active inflammatory CD represents a major paradigm shift in approach.^[2,3] Why it works has baffled the IBD scientific community who have tried to fit the effects of EEN into the currently held views on the pathogenesis of CD that are at best vague, summed up as 'a complex interplay between genetic predisposition, environmental factors, the gut microbiota and the host immune system'.^[4] The impact of EEN has not only been on the patients, but also on the research community who are earnestly searching for 'real food' dietary interventions that might emulate the efficacy of EEN without such a degree of restriction. This effort goes hand-in-hand with attempts to identify dietary factors that might lead to or protect from the development of CD, with the view that such factors might also be involved in perpetuating disease once it has developed. Specific dietary factors and/or dietary patterns, are reviewed in detail elsewhere.^[5] Contrary to such concepts, presumed protective factors (fibre and docosahexaenoic acid) are rarely present in EEN formula^[6] and components implicated by some in CD development or progression (such as emulsifiers and maltodextrin) are present in EEN formula,^[6] often in doses higher than that found in food.^[7] These somewhat contradictory concepts raise questions around our current understanding of the pathogenesis of CD.

The greatest limitation to achieving positive outcomes with EEN in patients with CD is lack of compliance and/or intolerance of the liquid feed. The corollary is that the rates of response or remission and of mucosal healing are very high (exceeding 70%) when used per protocol. Earlier experience suggested that children did better than adults^[8] and that those with ileal disease did better than those with colonic disease. However, more recent data (albeit mainly observational) indicate high remission rates are possible in adults^[9,10] and no evidence that disease phenotype influences remission rates.^[11] In routine adult clinical practice in both ambulatory and in-patients, adults do as well as children if they take the therapy per protocol. That is not to say that the mechanisms of action are homogeneous, but may differ according to the clinical scenario and phenotype. For example, the role of nutritional repletion (and subsequent restoration of normal growth patterns) may play a greater role in children than in adults. Most of the mechanistic data has derived from children since this is where this therapy first gained acceptance.

Hence, the aim of this narrative review was to critically examine the current available evidence regarding the putative mechanisms of action of EEN and to identify gaps that might be amenable to investigation. All data will be presented together, but will be identified if deriving from adult or paediatric populations.

Table 1. Changes in faecal microbiota in responding and non-responding paediatric patients with active Crohn's disease receiving exclusive enteral nutrition (EEN).
Study	Participants	EEN duration	Microbiota analysis	Changes in microbiota observed
Study	Participants	EEN duration	Microbiota analysis	Responders	Non-responders
Gerasimidis et al. (2014)	n = 15	8 weeks	TTGE Real-time quantitative PCR	Greater reduction in F prausnitzii Bacteroides/Prevotella group decreased (p < 0.05) E. coli tended to decrease
Leach et al. (2008)	n = 6; new diagnosis	Median 56 (range 46–57) d	PCR DGGE	Bacteroides/Prevotella change (Bray–Curtis % similarity) associated with reduction in clinical disease activity (PCDAI)
Dunn et al. (2016)	n = 10	12 weeks	16s rRNA gene sequencing	Species diversity tended to decrease Increased abundance of Akkermansia muciniphilia and Bacteroides	Species diversity tended to increase Increased abundance of Proteobacteria
Kaakoush et al. (2015)	n = 5; new diagnosis	Variable (8–12 weeks)	16s rRNA gene sequencing	Greater reduction in microbial diversity and OTUs
Diederen et al. (2020)	n = 43	6 weeks	16s rRNA gene sequencing	Higher proportional abundance • Genus: Roseburia, Faecalibacterium • Species: F prausnitzii, B adolescentis, Ruminococcus bromii	Higher proportional abundance • Escherichia coli • four Blautia OTUs

Abbreviations: DGGE, denaturing gradient gel electrophoresis; PCR, Polymerase chain reaction; PCDAI, Paediatric Crohn's Disease Activity Index; TTGE, Temporal temperature gradient gel electrophoresis.

Table 2. The impact of exclusive enteral nutrition (EEN) on immunological indices.
Study	Participants	Duration	Formula type	Findings
Yamamoto et al. (2005)	n = 28; adults Active disease	4 weeks	Elemental	Reduction in mucosal inflammatory cytokines Normalisation in the IL-1ra:IL-1 ratio (correlated with mucosal healing)
Yamamoto et al. (2007)	n = 40; adults Clinical remission (CDAI <150)	>1 year	Polymeric (nasogastric tube overnight) + low-fat diet daytime vs. no diet therapy	Mucosal IL-1β, IL-6 and TNF-α concentrations at 12 months: • significantly increased with time in untreated group • no change in EEN group
Fell et al. (2000)	n = 29; children Active disease	8 weeks	Polymeric	Downregulation mucosal inflammatory cytokines In colon and ileum: • IL-8 mRNA elevated pre EEN in colon only; reduced following treatment • Reduced Interferon-γ (IFN-γ) mRNA in ileum only • Reduction in interleukin-1β (IL-1β) mRNA in ileum and colon • 10-fold increase in TGF-β1 mRNA in ileum only
Rolandsdotter et al. (2019)	n = 6; children Active disease	6 weeks	Polymeric	Reduced overall expression mucosal cytokines No significant difference in mucosal cytokines profiles
Schwerd et al. (2016)	n = 15; children Active disease	3 weeks	Elemental or polymeric	In peripheral blood mononuclear cells: • reduced secretion of IL-6, IL-8, IL-1β and TH1-derived IFN-γ when stimulated with bacterial ligands • Enhanced capacity of IL-10 to suppress LPS-induced IL-6 • Increased FOXP31 Treg and gut-homing TH cells expressing α4β7-integrins In lamina propria, reduced Treg cells