Nutritional interventions for the treatment of IBD: current evidence and controversies

The role of diet in the pathogenesis of IBD

Crohn’s disease (CD) and ulcerative colitis (UC) are chronic intestinal inflammatory diseases of unknown aetiology, characterized by a recurrent inflammatory condition. The pathophysiology is multifactorial and complex, involving an inappropriate immune activation of the gut mucosa in genetically susceptible individuals, triggered by an altered composition of the gut microbiota. The incidence of inflammatory bowel disease (IBD) has increased worldwide in developed nations and more recently in developing countries.^1,2 This rapid increase in the incidence of IBD over the last half-century, particularly in developing countries, clearly points to the role of changing environmental factors intrinsically implicated in disease development. ¹ Genes cannot change within such a short time frame, thus disease susceptibility remains almost identical over several generations. The differing role of genetics and the environment in disease development is clearly reflected by epidemiological data on immigrants from low- to high-IBD-incidence regions: where, the second generation of immigrants has the same risk of developing IBD as the local population living for generations in the same area. ³ A western lifestyle, including changes in dietary habits, urbanization and industrialization, has been proposed as one explanation for this worldwide increase in IBD. ⁴

Among environmental factors, diet is widely thought to play a pivotal role in the development of IBD. Epidemiological studies suggest a western diet, high in fat and protein content and low in fruits and vegetables, as a possible explanation for the recent increase in IBD. ⁵ It is important to address the question of whether there is a particular window of vulnerability/susceptibility to developing IBD with regard to a particular dietary exposure during infancy, childhood or adulthood or whether the simple and prolonged exposure to certain food components is a risk factor per se. Several large longitudinal studies have demonstrated an association between a reduced risk of developing IBD and a diet rich in fruits and vegetables, whereas consuming high amounts of animal fats and refined sugar is associated with an increased IBD risk.^6–8 Epidemiological studies indicate a higher consumption of red and processed meat, dietary fat and low levels of vitamin D are associated with an increased risk of developing IBD.^7,9,10 During recent decades, a shift in the composition of the daily diet occurred with a greater proportion of processed, prepackaged foods high in fat and carbohydrates, and low in dietary fibre. The evolutionary hypothesis has recently emerged developing the idea that the human digestive tract is insufficiently evolved to handle foods resulting from modern agricultural techniques: the exposure of the human digestive tract to foods that were not present at the time of human evolution may result in modern diseases. One example largely supporting this hypothesis is the use of emulsifiers in current foods. Emulsifiers are added to most processed foods to enhance taste, aid texture and extend shelf life. Chassaing and colleagues ¹⁰ demonstrated that emulsifiers induce a chronic intestinal inflammatory state promoting the development of chronic colitis in susceptible mice. In this experimental work, the intestinal microbiota were critical in triggering inflammation, since germ-free mice did not develop colitis upon exposure to emulsifiers. Another example is the food additive carrageenan, largely used in western foods to improve the texture. The proinflammatory character of carrageenan is well known and a recent small randomized trial in adult patients with UC ¹¹ showed a relapse in three out of five patients exposed to carrageenan compared with none out of seven patients on a carrageenan-free diet.

The triple interaction between diet, intestinal microbiota and the host

The effects of nutrition and diet on the host are multiple: food ingredients can directly interact with epithelial cells or under certain conditions with the immune system, as well as indirectly via gut microbiota. Gut microbiota form a complex and dynamic system with a steady state, which can be perturbed by many environmental factors, including diet. A homeostatic balance of the host–bacteria relationship is important and vital for a normal health process. An imbalanced intestinal microbiota termed ‘dysbiosis’ has been repeatedly seen in IBD and is now recognized as a key factor in gut inflammation. ¹² Sokol and colleagues ¹³ showed a significant decrease in the proportion of the Clostridium leptum phylogenetic group in patients with colonic CD. These results were confirmed by a metagenomic approach, revealing a restriction in biodiversity depending on bacteria belonging to the Firmicutes phylum (C. leptum and Clostridium coccoides) with a decrease in the proportion of bacteria belonging to the C. leptum phylogenetic group.^14,15 This dysbiosis is characterized by a high instability of the microbiota over time, the presence of approximately 30% of unusual bacteria, a marked increase in mucosal bacterial concentration, and a restriction in biodiversity regarding the Firmicutes phylum. Within this group, a decrease in C. leptum has been shown and particularly its major representative, Faecalibacterium prausnitzii. ¹²

Dietary habits are one of the main factors contributing to the diversity of human gut microbiota. Diet may directly affect the microbiome, thereby modifying the interaction with the host. ¹⁶ A study from De Filippo showed that gut microbiota from children with high fibre intake living in a rural African village in Burkina Faso, an environment resembling that of Neolithic subsistence farmers, is significantly different from the microbiota of children living in the urban western world with an enrichment in Bacteroidetes and particularly Prevotella spp. and Xylanibacter spp., but fewer Firmicutes in African compared with European children. ¹⁷ A fibre and plant-derived polysaccharide-rich diet in children and adults is associated with human gut microbiota enriched in Bacteroidetes compared with Firmicutes.^16,17 High-fat diets were shown to increase bowel permeability through dietary-induced changes in the gut microbiota. ¹⁸ High-fat diets also worsen dextran sodium-sulphate-induced (DSS) colitis in mice, possibly by increasing colonic epithelial nonclassical natural killer T cells and reducing regulatory T cells. ¹⁹ Llewellyn and colleagues, ²⁰ using a mouse model of colitis to control the host genotype and the gut microbiota, tested the effect of more than 40 dietary combinations on the severity of colitis induced by DSS. They found that faecal microbiota varied with diet and that the concentration of protein and fibre had the strongest effect on the development of colitis.

The gut microbiota continually produces a diverse repertoire of metabolites from fermentation of both exogenous and endogenous sources. ²¹ These metabolites serve as important signals that contribute to the regulation of the host physiology and maintenance of health. A role for dietary fibre in the predisposition to IBD appears to have particularly compelling biological plausibility. For example, the fermentation of nondigestible carbohydrates stimulates the growth of bacterial short-chain fatty acid (SCFA) producers. SCFAs mediate a number of important functions for the host, including use as an energy source by intestinal epithelial cells, and a variety of anti-inflammatory properties in T cells, regulatory T cells, neutrophils, and macrophages, where they affect migration, cytolytic activity, cytokine production, and epigenetic regulation of gene expression. ²² In addition, fibre plays a vital role in the maintenance of normal intestinal barrier function.

Diet may also have an impact on the mucosal layer of the digestive tract. Recent studies have demonstrated that microbiota can influence the properties of the colonic mucus layer. ²³ Some bacteria possess various enzymes allowing them to degrade and metabolize specific glycans of the intestinal mucosal layer. Desai and colleagues ²⁴ demonstrated that low-fibre diet promotes expansion and activity of colonic mucus-degrading bacteria and that fibre-deprived gut microbiota promote aggressive colitis by an enteric pathogen.

Another mechanism of dietary interaction with the host is via a direct effect on the immune system, where dietary antigens by themselves may trigger an immune response. The typical western-style diet, high in sugar and saturated fats and low in fibre, can lead to systemic low-grade inflammation, as described in obesity. ²⁵ In animal models, the consumption of milk-derived saturated fats alters bile acid composition, allowing for a bloom of sulphate-reducing bacteria, which in turn can produce greater amounts of the potentially mucosal toxic hydrogen sulphide and production of helper T cell (Th1) proinflammatory cytokines. ²⁶

Because dietary antigens, along with bacterial antigens, are the most common types of luminal antigen, it is reasonable to suppose that dietary factors may play an important role in the pathogenesis of IBD, possibly by interacting with both gut microbiota and the mucosal immune system. Current IBD therapies predominantly target the pathological immune responses rather than any potential causal factors on top of the inflammatory cascade. Given that diet is a modifiable environmental risk factor, it has become an attractive target both for the prevention and treatment of IBD.

Oral diets and IBD

Many different types of diets have been proposed in the treatment of IBD. They are all based on the exclusion of some dietary components. Intense research is ongoing to improve nutritional approaches given the limitation of EEN due to its exclusivity character not allowing any additional food. The study of Johnson and colleagues ⁴⁴ confirmed that partial enteral nutrition (allowing an unrestricted free diet up to 50% of daily caloric intake) is not effective in inducing complete remission in CD. However, the recent innovative approach of Rotem Sigall-Boneh and Professor Arie Levine offers a very promising alternative to strict EEN. This new diet, called the CD exclusion diet (CDED) is based on the idea of excluding potentially ‘harmful’ or proinflammatory food ingredients. The first report in 2014 showed a high potential to induce clinical remission when combining enteral nutrition with the CDED ⁴⁵ : all patients received one half of calories from enteral nutrition and one half from an exclusion diet avoiding gluten, dairy products, gluten-free baked goods and breads, animal fat, processed meats, products containing emulsifiers, canned goods, and all packaged products. Clinical remission was obtained in 33 of 47 (70%) of children/adolescents/young adults and normalization of C-reactive protein occurred in 21 of 30 (70%) patients in remission. In this study, seven patients were included who did not take any formula, but followed CDED and six of seven achieved clinical remission with the solid food diet alone. This raises the question of whether the supplementary milk formula is indispensable; it seems rather that it serves as a guarantee for a balanced diet, avoiding deficiencies. A subsequent clinical trial randomizing n = 78 patients to EEN or partial enteral nutrition plus CDED showed a comparable efficacy with 30/40 patients in remission at week 6 receiving CDED versus 20/34 receiving EEN (p = 0.38). While efficacy was similar, tolerance was higher in the CDED group. ⁴⁶ So far, no data on mucosal healing with CDED are reported, but studies with follow-up endoscopy in patients on CDED are underway. The first data of Professor Arie Levine’s studies on CDED indicate that patients coming into remission on CDED induction therapy stay in prolonged remission while following phase II and III of the exclusion diet. Thus, there are clear indicators to believe that enteral nutritional therapy might also play a significant role as maintenance therapy for CD. The initial preliminary analyses of Professor Levine’s group indicate that there is a clear difference in the modulation of the intestinal microbiota in patients coming into and maintaining remission on CDED compared with those failing remission. ⁴⁶

A somewhat different approach for inducing remission in patients with CD by the sole use of diet was recently reported by the Scottish group led by Professor Richard Russell with a diet called CD-TREAT. ⁴⁷ This diet is based on an idea to mimic, with solid food, the composition and effects of enteral nutrition, particularly on the intestinal microbiota. In a first step, the authors compared the effect of CD-TREAT with EEN on the microbial composition and function (including short-chain acid and sulphite production) in healthy adult volunteers. A total of 25 healthy volunteers were randomized to receive EEN or CD-TREAT first and after a 2-week washout they were crossed over to the other treatment arm allowing individual comparison of changes in faecal microbiota and metabolome. As excepted, tolerance of CD-TREAT was superior to EEN, but the effects on gut microbiota composition and metabolic functions were comparable, encouraging the team after obtaining animal data confirming the anti-inflammatory potential of CD-TREAT to propose it to five children with active CD. After a course of 8 weeks, four of five children responded to CD-TREAT, with three children achieving clinical remission. Only one patient normalized faecal calprotectin on CD-TREAT, the four other children showed a decrease from a mean 1960 ± 1104 mg/kg at baseline to a mean 1042 ± 776 mg/kg at 8 weeks.

One of the most studied exclusion diets for the treatment of IBD is the Specific Carbohydrate Diet (SCD). Initially, it was developed for the treatment of coeliac disease, ⁴⁸ but it became very popular in IBD due to several impressive lay reports indicating the potential to treat various diseases, including UC. The principle of the SCD is to remove grains including wheat, barley, corn, and rice; added sugar is limited to honey and also most milk products are restricted except for fully fermented yogurts. The rationale of the SCD is based on the notion that polysaccharides and complex sugars are poorly absorbed in the intestinal tract causing potentially bacterial (and possibly fungal) overgrowth with subsequent inflammation of the intestinal and colonic mucosa. Suskind and colleagues reported clinical remission in seven children with CD after a SCD diet of several months (average 14.6 ± 10.8 months) with a significant improvement of inflammatory markers. ⁴⁹ An additional retrospective study of the same group ⁵⁰ confirmed in 20 children with CD and 6 children with UC a marked reduction in disease activity scores under SCD, in line with the observation of 50 adult patients with IBD responding favourably to prolonged SCD. These data were obtained by a patient reported 3-day dietary survey. ⁵¹ Cohen and colleagues ⁵² reported that 8 of 10 children/adolescents with CD responded to a 12-week cycle with SCD; 1 was not able to take the diet, while 1 patient with colonic CD experienced a relapse. In the following study, seven patients continued the diet for 52 weeks. Since these first observations, several studies have been performed, analysing the potential of SCD to control inflammation in children and adults with IBD. Suskind and colleagues ⁵³ showed, in a prospective study on 12 children with mild to moderate IBD (CD and UC), that the addition of SCD over a 12-week period to ongoing medication allowed remission to be reached in 8 of 12 patients; 2 patients did not respond to SCD and two patients were not able to maintain the diet. Microbial analyses of these patients confirmed a marked difference between baseline and after 12 weeks of SCD. Additional retrospective, studies on CD and UC, including one online survey, confirmed the perception of a clinical benefit and the potential of SCD to reduce inflammation and its efficacy inducing and maintaining remission over several months.^50,54,55 While weight loss is a concern when utilizing SCD, many academic IBD centres have developed active dietary programs supporting patients on SCD.^50,51,56 In the study of Cohen and colleagues, ⁵² repeat endoscopic evaluations were available, indicating the potential of SCD to induce mucosal healing, as shown for two patients and also in the report by Miller and colleagues. ⁵⁷ Another study ⁵⁸ failed to demonstrate achieving mucosal healing despite a positive clinical evolution in patients on a modified SCD (integrating some initially excluded foods, such as potatoes, rice, quinoa or oats). Since adherence to SCD, as to other exclusion diets, can be a challenge and modification of SCD might compromise the clinical anti-inflammatory effects, well-designed and precise prospective randomized trials are needed to further affirm this promising treatment option.

The Palaeolithic diet ⁵⁹ is based on the idea that the human intestinal tract is not sufficiently evolved to digest food from modern agriculture and thus, the exposure to foods not present at the time of human evolution may cause modern diseases, such as IBD. In common with other diets, the Palaeolithic diet excludes many food ingredients and privileges the intake of lean, nondomesticated meats and noncereal plant-based foods (i.e. fruits, roots, legumes, and nuts). So far, no data exist on the role of the Palaeolithic diet in the treatment of IBD, except of rare and isolated rather positive case reports.

With time, up to one third of patients with IBD will develop functional gastrointestinal symptoms, sharing similarities with those of irritable bowel syndrome. ⁶⁰ With the development of better strategies to deal with functional symptoms in patients with IBS, some were extrapolated to patients with IBD. An important strategy to improve symptoms is diet. The most common diet used for IBS is a gluten-free diet. This diet may be associated with a better digestive wellbeing according to patients but controlled trials are missing to conclude on the usefulness of this restrictive diet. The second diet used successfully for patients with IBS is a low-FODMAP (Fermentable Oligo-, Di-, Mono-saccharides And Polyols) diet. The diet is based on the reduced intake of indigestible and slowly absorbed carbohydrates that may induce symptoms through luminal distension and mechanoreceptor stimulation by virtue of their osmotic effects and fermentation. This is the basis for the lactose-reduced diet in patients with lactose malabsorption and the low-FODMAP diet, in which all short-chain carbohydrates are reduced. The recent prospective study of Cox and colleagues ⁶¹ demonstrated that patients with quiescent IBD scored higher for quality-of-life scores while on a low-FODMAP diet than patients on a control diet, whereas the IBS scores did not significantly differ in the same study. In the same line, the study of Halmos and colleagues ⁶² showed no effect of reducing inflammatory parameters on a low-FODMAP diet. A recent meta-analysis ⁶³ is largely supportive that a low-FODMAP diet is beneficial for reducing gastrointestinal symptoms in patients with quiescent IBD; there was a significant improvement in symptoms like diarrhoea, abdominal bloating, abdominal, fatigue and nausea. However, a major drawback for patients with IBD might be a potential reduction of the abundance of anti-inflammatory bacteria and thus the risk of aggravating dysbiosis on a prolonged low-FODMAP diet.^61,62

A short pilot study reported symptomatic improvement through excluding foods with high immunoglobulin (Ig)G4 levels (mostly eggs and beef). ⁶⁴

A semi-vegetarian diet (lacto-ovo-vegetarian diet) has been shown to maintain a longer clinical remission in patients with CD compared with an omnivorous diet with a remission rate at 2 years of 94% versus 33%. ⁶⁵ This highlights the role for ‘good’ fibres in the diet in order to enrich the microbiota by giving prebiotics and substrates to protective bacteria. Recent studies have demonstrated that a diet high in fibre is beneficial to both patients with UC and CD.^66,67

What clinicians should tell their patients regarding diet