ESPEN guideline: Clinical nutrition in inflammatory bowel disease

The rising incidence of IBD in Western countries has generally predated that in developing nations, supporting the hypothesis that 'Westernization' of our lifestyle has led to the increased incidence of IBD. Smoking, antibiotic use, and diet are potentially reversible risk factors for IBD. Multiple dietary components may impact on the resident flora, generating dysbiosis diminishing or damaging the mucus layer, may increase intestinal permeability or increase the ability of pathological microbiota to adhere to epithelial cells or translocate across the epithelial barrier. For example, in a recent study it has been shown that western diet induces changes in the composition of gut microbiota, alters host homeostasis and promotes an unfavourable gut colonisation in genetically susceptible mice [].

Many studies have evaluated the effect of diet on the risk of developing IBD. However most of them are retrospective case–control studies. In 2011 Hou and al. published the first systematic review entitled “Dietary Intake and Risk of Developing IBD” []. They used guideline-recommended methodology to evaluate the association between pre-illness intake of nutrients (fats, carbohydrates, protein) and food groups (fruits, vegetables, meats) and the risk of subsequent IBD diagnosis. Nineteen studies were included, encompassing 2609 IBD patients (1269 with CD and 1340 with UC), and over 4000 controls. The main results of this systemic review are the following:

• •
  There is an increased risk of developing UC with high intake of total fat, PUFAs, omega-6 fatty acids, and meats,
• •
  There is an increased risk of CD with high intake of PUFAs, omega-6 fatty acids, saturated fats, and meat.
• •
  There is a decreased risk of CD, but not UC, with high intake of dietary fibre and fruits. A consistent association was shown between high dietary fibre and decreased risk of CD, with the protective effect observed to be statistically significant in those consuming more than 22.1 g/d. The review also observed that a high intake of fruit is associated with a 73–80% decreased risk of CD. This association was confounded by dietary fibre intake and the fact that a diet high in fruits may conversely be low in fats and meats.
• •
  There is no consistent association between total carbohydrate intake and IBD risk, even in studies reporting intake greater than double the recommended daily intake.

Some important studies from established prospective cohorts [the Investigation into Cancer and Nutrition (EPIC) cohort and the Nurses' Health Study I and II cohorts], have been recently published and bring additional and important new insights.

Fibre, fruit and vegetables: In a large prospective cohort study including 170,776 female registered nurses followed over 26 years, 269 incident cases of CD and 338 cases of UC were identified []. Compared to women with the lowest energy-adjusted fibre intake, intake of fibre in the highest quintile (median 24 g per day) was associated with a significant reduction in risk of CD [hazard ratio (HR) 0.59, 95% confidence interval (CI) 0.39–0.90] but not UC. Interestingly, this association seemed specific for fibre from fruits in particular, and only to a lesser degree from vegetables and cruciferous vegetables. No association was identified between intake of fibre from other sources such as cereals, whole grains, or legumes. This association was also slightly stronger with respect to small bowel as opposed to colonic CD.

In a recent meta-analysis including a total of 14 case–control studies [], consumption of vegetables was negatively associated with the risk of UC (OR = 0.71, 95% CI 0.58–0.88, n = 9 studies), but not with CD (OR = 0.66, 95% CI 0.40–1.09, n = 8 studies). Higher consumption of fruit was negatively associated with the risk of UC (OR = 0.69, 95% CI 0.49–0.96, n = 8 studies) and CD (OR = 0.57, 95% CI 0.44–0.74, n = 10 studies). On subgroup analysis the intake of vegetables was negatively associated with the risk of CD in studies carried out in Europe (OR = 0.36, 95% CI 0.23–0.57), but not in Asia (OR = 1.00, 95% CI 0.50–2.03).

Dietary fat: Among the 170,805 women enrolled in the Nurses' Health Study the effect of energy-adjusted cumulative average total fat intake, as well as specific types of fat and fatty acids, on the risk of CD and UC was examined using Cox proportional hazards models adjusting for potential confounders []. Cumulative energy-adjusted intake of total fat, saturated fats, unsaturated fats, n-6 and n-3 polyunsaturated fatty acids (PUFA) were not associated with risk of CD or UC. However, greater intake of long-chain n-3 PUFA was associated with a trend towards lower risk of UC (Hazard ratio (HR) 0.72; 95% CI 0.51–1.01). In contrast, high long-term intake of trans-unsaturated fatty acids was associated with a trend towards an increased incidence of UC (HR 1.34, 95% CI 0.94–1.92).

In the EPIC study, 229,702 participants were recruited from nine European centres between 1991 and 1998 []. At recruitment, dietary intakes of DHA and fatty acids were measured using validated food frequency questionnaires. In a nested case–control analysis, each participant who developed incident UC (n = 126) was matched with four controls. The highest quartile of intake of linoleic acid was associated with an increased risk of UC (odds ratio (OR): 2.49; 95% CI: 1.23 to 5.07, p = 0.01) with a significant trend across quartiles (OR 1.32 per quartile increase (95% CI: 1.04 to 1.66; p = 0.02 for trend). In another nested case–control analysis of the EPIC study [], each participant who developed incident CD (n = 79) was matched with four controls. All higher quintiles of DHA intake were inversely associated with development of CD; the highest quintile had the greatest effect size (OR 0.07; 95% CI 0.02–0.81). The OR trend across quintiles of DHA was 0.54 (95% CI 0.30–0.99). Including BMI in the multivariate analysis, due to its correlation with dietary fat showed similar associations. There were no associations with the other dietary fatty acids studied.

Looked at from an alternative perspective in nearly 200 children with a new diagnosis of CD, Costea et al. again concluded that a high omega-6:omega-3 ratio in the diet predisposes to the condition (odds ratio of up to 3), but that this is the case only for those with specific polymorphisms of the CYP4F3 and FADS2 genes []. The two genes code for a leukotriene B4 inhibitor and for enzymes in PUFA metabolism respectively and further support an interaction between nature and nurture in IBD.

It is also possible (and of relevance to nutrition when it is used therapeutically) that it is not only the fats themselves that are important, but additional agents employed to keep them in forms that are aesthetically acceptable. The emulsifiers used in commercially prepared foods may be implicated in this regard, with at least one (polysorbate 80) having a proposed specific mechanism as it increases bacterial translocation across the intestinal epithelium [].

Vitamin D: Khalili et al., using the Nurses' Health Study cohort, demonstrated a lower risk for both CD (HR 0.48, 95% CI 0.30–0.77) and UC (HR 0.62, 95% CI 0.42–0.90) in women who were residing in southern latitudes at age 30, compared to those residing in northern latitudes []. In a prospective cohort study of 72,719 women (age, 40–73 y) enrolled in the Nurses' Health Study, women completed an assessment of diet and lifestyle, from which a 25-hydroxy vitamin D [25(OH)D] prediction score was developed and validated against directly measured levels of plasma 25(OH)D []. During 1,492,811 person-years of follow-up 122 incident cases of CD and 123 new cases of UC were documented. The median predicted 25(OH)D level was 22.3 ng/mL in the lowest, and 32.2 ng/mL in the highest quartiles. Compared with the lowest quartile for vitamin D levels, the multivariate-adjusted HR for CD was 0.54 (95% CI: 0.30–0.99) in the highest quartile for vitamin D, and 0.65 (95% CI, 0.34–1.25) for UC. Compared with women with a predicted 25(OH)D level less than 20 ng/mL, the multivariate-adjusted HR for UC was 0.38 (95% CI, 0.15–0.97) and a non-significant 0.57 for CD (95% CI, 0.19–1.70) for women with a predicted 25(OH)D level greater than 30 ng/mL. There was a significant inverse association between dietary and supplementary vitamin D and UC, and a non-significant reduction in CD risk.

Zinc: There has been limited examination of the role of micronutrients in IBD pathogenesis. Dietary zinc is promising as a risk factor and may influence risk of IBD through effects on autophagy, innate and adaptive immune response and maintenance of the intestinal barrier. In a recent study concerning zinc intake and incidence of IBD, data from 170,776 women from the Nurses Health Study I and Nurses Health (using semi-quantitative food questionnaire) were presented. There were 269 incident cases of CD and 338 of UC []. Zinc intake ranged from a median of 9 mg/day in the lowest quintile to 27 mg/day in the highest quintile. Compared to women with the lowest quintile of intake, the multivariate hazard ratios (HR) for CD were 0.92 (95% CI, 0.65–1.29) for the second quintile of intake, 0.60 (95% CI, 0.40–0.89) for the third quintile, 0.57 (95% CI, 0.38–0.86) for the fourth quintile, and 0.74 (95% CI, 0.50–1.10) for the highest quintile (p for trend = 0.003). Compared to individuals with intake of zinc less than the recommended daily allowance (8 mg/day), those with an intake of 8–16 mg/day (HR 0.69, 0.44–1.08) and >16 mg/day (HR 0.52, 0.32–0.86) had a reduced risk of CD. The association was stronger for dietary zinc (HR 0.63, 95% CI: 0.43–0.93), comparing extreme quintiles, than for zinc intake from supplements. In conclusion, in two large prospective cohorts of women, intake of zinc was inversely associated with risk of CD but not UC.

Dietary pattern: Within the prospective EPIC programme, a nested matched case–control study was performed among 366,351 participants with IBD data, which included 256 incident cases of UC and 117 of CD, and 4 matched controls per case []. Dietary intake was recorded at baseline from validated food frequency questionnaires. Incidence rate ratios for the development of UC and CD were calculated for quintiles of the Mediterranean diet score, and a posteriori dietary patterns were produced from factor analysis. No dietary pattern was associated with either UC or CD. Specifically there were no associations with a Mediterranean diet and either condition. However, when excluding cases occurring within the first 2 years after dietary assessment, there was a positive association between a “high sugar and soft drinks” pattern and UC risk (incidence rate ratios for the 5th versus the 1st quintile: 1.68 (1.00–2.82). When considering the foods most associated with the pattern, high consumers of sugar and soft drinks were at higher UC risk only if they had low vegetable intakes.

Other micronutrients, microparticles and the unintentional inclusion of trace metals in the diet, such as by the swallowing of toothpaste, have been explored and there are no robust data to indicate important effects on IBD pathogenesis (reviewed by Andersen et al. []).

In conclusion, the external environment offers particular promise as a modifiable risk factor for both incident disease and for outcomes in those with established disease []. Many concordant results suggest that a diet rich in fruits and vegetables in n-3 fatty acids and low in n-6 fatty acids is associated with a decreased risk of developing CD or UC. Interesting new data suggest that a diet rich in vitamin D and zinc may also protect against CD but not UC. Rigorous randomized controlled trials examining the effect of dietary factors are required to establish or refute the role of these factors in achieving and maintaining disease remission.

An early case control study conducted in 9 countries included 499 patients to investigate childhood factors predicting IBD yielded no significant differences between patients and controls in the frequency of breast feeding, cereal consumption, sugar added to milk in infancy, and other dietetic factors []. This finding was confirmed in a German study []. In contrast, an Italian study indicated that lack of breastfeeding is associated with an increased risk of UC (OR = 1.5; 95% CI: 1.1–2.1) and CD (OR = 1.9; 95% CI: 1.1–3.3) []. Systematic reviews from 2004 and 2009 concluded strongly in favour of breastfeeding [, ] and subsequent studies have reinforced this interpretation. A case–control study from New Zealand reported that breastfeeding was protective against IBD (CD OR 0.55 [0.41–0.74], UC OR 0.71 [0.52–0.96]) with a duration-response effect []. Comparable data were reported from a Danish cohort study, in which breastfeeding for >6 months decreased the odds of IBD (OR, 0.50; 95% CI, 0.23–1.11) []. More recently still, 2 further publications confirmed this relationship, one from the US and another from Asia–Pacific. The US study was a single centre study in which the relation between breastfeeding and requirement for disease-related surgery in 333 CD and 270 UC patients was examined. Among those with CD, being breastfed was associated with reduced risk of CD-related surgery (34% vs. 55%), while none of the early life variables influenced disease phenotype or outcome in UC []. The Asia–Pacific study included 442 incident IBD cases from eight countries in Asia and Australia and 940 controls. In a multivariate model, being breastfed for >12 months decreased the odds for CD (aOR 0.10; 95% CI 0.04 to 0.30) and UC (aOR 0.16; 0.08 to 0.31) in Asians [].

Breastfeeding for around six months is desirable in all infants []. Regarding longer periods of breastfeeding, current European recommendations suggest that breastfeeding is continued as long as mutually desired by both mother and infant []. In summary, the majority of the literature (and in particular the more recent publications) supports the importance of breastfeeding as a protective factor in early childhood regarding the development of IBD.

Adults with IBD are at increased risk of malnutrition, with deficits more common in patients with CD than UC []. Obese patients may have covert deficits in lean mass which may be unmasked by tools such as skinfold thickness measurement. Patients with active IBD, particularly those whose disease is poorly responsive to medical therapy, are at highest risk of poor nutrition. In adults, risk of malnutrition can be assessed with validated screening tools [].

Malnourished patients with IBD are more likely to be hospitalised following emergency department attendance [] and are more likely to be admitted to hospital due to infection []. In hospitalised patients malnutrition is an independent risk factor for venous thromboembolism [], non-elective surgery [], longer admission [, ] and increased mortality [].

Pragmatically optimising nutrition status may improve outcomes for patients with IBD therefore it is logical to screen for, and manage, undernutrition using an appropriately trained multidisciplinary team.

Malnutrition in children: Malnutrition in childhood Crohn's is common at diagnosis and may persist despite disease treatment []. Children with UC are also at risk of poor nutrition but nutritional deficits may not be immediately obvious on assessment of just height and weight []. Although a variety of screening tools exists, the tools have poor ability to discern different levels of nutrition risk for children with IBD []. Poor nutrition in childhood IBD contributes to disrupted pubertal development and impaired growth velocity which may lead to short stature in adulthood.

Malnutrition plays a role in the pathogenesis of IBD, in its clinical presentation and in disease treatment and outcome. As in adults, the mechanisms involved include limited food intake, malabsorption of nutrients, and increased nutrient losses. With specific drugs (sulfasalazine, methotrexate, steroids) it can include interactions between these drugs and nutrients.

Of particularly importance in paediatric IBD is growth failure, which is the result of a combination of inflammation and chronic malnutrition []. Growth failure is seen in 15–40% of children with IBD [, ]. Both growth failure and delay of puberty are more common in Crohn's than in UC. Despite greater disease awareness, growth failure is still found to precede the diagnosis of Crohn's by many years in a high proportion of patients. This may have an adverse effect on the final height of these patients, who commonly fail to reach their final predicted height: short stature (final height below 5th percentile) is present in up to 30% of Crohn's patients [].

Iron deficiency is particularly common in paediatric IBD, while other deficiencies include folic acid, zinc, magnesium, calcium, vitamins A, B12, D, E, and K []. A detailed discussion of nutritional assessment is beyond the scope of these guidelines, however, a careful account of nutrition intake, anthropometric measurements, including history of growth with plotting of previous measurements of weight and height and assessment of growth rate are essential. Laboratory work up to identify and treat nutrient deficiencies is also essential.

For clarity this question can be formulated in two ways; firstly do patients with IBD have an altered energy requirement compared to healthy individuals, and secondly do energy requirements vary with disease activity. It is also worth noting that an individual patient's daily energy requirement includes their resting energy expenditure (REE), which includes the energy cost of depositing tissue/growth, energy expended in physical activity, and dietary induced thermogenesis. An important consideration highlighted in paediatric data is how to adjust for differences in energy expenditure attributable to body size: patients with greater mass have greater REE. This effect may not be fully negated by expressing REE per unit of mass or lean mass, and alternative analyses have been proposed [, , ].

There are relatively few studies examining energy expenditure in patients with UC and all studies are of only small numbers of patients. There may be an increase in metabolic activity at times of acute severe colitis compared to remission in adults [, ] which is understandable considering that systemic disturbance (fever and tachycardia) is common. However, an increase in REE is likely to be offset by reduction of physical activity. Significant reduction in dietary intake is common in acute colitis and may result in negative energy balance []. Inconsistent results about changes in resting energy expenditure are found for milder disease activity and for children.

One single study has measured total energy expenditure in adults with CD and recorded normal values []. Comparison between other studies of resting energy is hampered by differing presentation of data. However, measured REE has consistently been found to be similar to predictive equations based on weight in adults [, ] or children [, , , ]. Measured REE/kg in adult patients has been found to be higher than [] or the same as [] that measured in healthy controls. However, this could be due to inadequate consideration of body size and the relative proportions of tissues of differing metabolic activity. REE does not appear to be raised in patients with weight loss, but decreased nutrient intake and malabsorption has been shown in these patients [, ]. No consistent association between CD activity and REE in adults has been demonstrated. In children with Crohn's, measured REE has not been demonstrated to be significantly different in children before and after infliximab (anti-TNF) [, , ] and no consistent association has been found between REE/kg FFM and markers of disease activity [].

In summary, patients with IBD do not have an increased energy expenditure as a direct result of their disease and predictive equations are suitable for estimating requirements. Dietary intake may be inadequate to meet even normal requirements particularly during periods of disease activity which may lead to weight loss. Measurement of REE by indirect calorimetry could be used in troublesome cases.

Patients with IBD develop a relative reduction in lean mass and increase in adiposity over time. This may occur due to chronically poor dietary intake, increased rates of protein turnover and gut loss of nutrients during phases of active disease or from the effect of disease treatments. Corticosteroids increase net loss of protein in children [] and adults [] with Crohn's. In contrast administration of elemental or polymeric feed as treatment of Crohn's or as adjunctive nutrition support results in reduction of proteolysis and acquisition of lean tissue in children and adults [, , ]. In children with active CD one study examined the reduction in protein turnover resulting from treatment with Infliximab and demonstrated improved protein metabolism in patients receiving parenteral nutrition both before and after infliximab treatment [].

Monitoring of anthropometry provides insight into which patients develop relative deficits in lean mass and therefore would benefit from nutritional supplementation. There is no good evidence that the daily protein needs of IBD patients differ from those of healthy controls, but as discussed elsewhere poor appetite and restricted dietary intake is commonplace. In patients receiving steroids and gut rest, enteral tube feeding may provide beneficial effects on protein turnover without deleterious consequences on disease activity.

There is no good evidence that the daily protein needs of IBD patients in remission differ from those of healthy controls. Provision of 1 g protein for each kilogram of body weight is therefore reasonable. However in active inflammation the proteolytic, catabolic response justifies an increase in provision to 1.2–1.5 g/kg bodyweight [, ].

Patients with IBD are vulnerable to micronutrient deficits due to gut loss from diarrhoea and inadequate dietary intake from anorexia accompanying disease activity. At times when nutrition support is offered then multivitamin and micronutrient supplements should also be offered to ensure an appropriately balanced nutritional intake.

When interpreting blood results of micronutrients and trace elements it is important to consider that many serum values, or markers of status, are positive or negative acute phase reactants; Serum levels rise or fall, as part of the inflammatory response; for example ferritin, and copper increase but folate, selenium and zinc decrease in inflammation []. In light of this some authors have examined micronutrient status in patients in clinical disease remission and found deficits of a variety of micronutrients [, ]. Furthermore, deficits may be present even in apparently well nourished individuals []. These observations highlight the need for routine monitoring (perhaps annually) to screen for deficiency. A daily multivitamin supplement may correct most deficiencies but is no guarantee of adequacy, even over the long term; iron, zinc and Vitamin D are likely to require specific replacement regimens []. Poor compliance, particularly in adolescents, is common with multivitamin supplements and patient education about the rationale behind their use is important [].

Consequences of deranged micronutrient status include anaemia, impaired linear growth and poor bone health. Recent research has focused on Vitamin D; it and its receptor may have some immunomodulatory properties, which further highlights the need for specific attention to micronutrient status in patients with IBD.

Anaemia is considered the most frequent extraintestinal manifestation of IBD, usually complicating the course both in UC and Crohn disease (CD). Prevalence rates of anaemia in IBD vary widely from 6 to 74% []. Anaemia is reported more frequently in hospitalized patients with IBD and occurs more frequently in CD than in UC []. In IBD patients anaemia increases, morbidity, rate of hospitalization, medical costs and deaths [, ]. In the majority of cases, IBD-associated anaemia represents a combination of chronic iron deficiency and anaemia of chronic disease []. The currently used WHO definition of anaemia (Table 4) applies also to patients with IBD [].

All patients with IBD regardless of their age should be assessed for the presence of anaemia []. The major forms of anaemia in IBD are iron deficiency anaemia (IDA), anaemia of chronic disease (ACD) and anaemia of mixed origin [ECCO Anaemia Statement 1A]. Diagnostic criteria for iron deficiency depend on the level of inflammation. For laboratory screening, complete blood count, serum ferritin, and C-reactive protein [CRP] should be used [ECCO Anaemia Statement 1B]. For patients in remission or mild disease, measurements should be performed every 6–12 months. In outpatients with active disease such measurements should be performed at least every 3 months [ECCO Anaemia Statement 1B]. In patients without clinical, endoscopic, or biochemical evidence of active disease, serum ferritin <30 μg/L is an appropriate criterion for the diagnosis of IDA. In the presence of inflammation, a serum ferritin up to 100 μg/L may still be consistent with iron deficiency [ECCO Anaemia Statement 1D]. In the presence of biochemical or clinical evidence of inflammation, the diagnostic criteria for ACD are a serum ferritin >100 μg/L and transferrin saturation <20%. If the serum ferritin level is between 30 and 100 μg/L, a combination of true iron deficiency and ACD is likely [ECCO Anaemia Statement 1E].

Iron supplementation is recommended in all IBD patients, whatever their age, when iron-deficiency anaemia is present [ECCO Anaemia Statement 2A]. Quality of life improves with correction of anaemia, and this improvement is independent of clinical activity []. The decision to supplement iron in patients without anaemia is more controversial and will depend on the patients' history, symptoms and individual preferences. Although there is evidence of benefit in treating iron deficiency without anaemia in other conditions such as chronic fatigue and heart failure, such evidence is not yet available in the context of IBD []. In a recent meta-analysis of randomized controlled trials comparing intravenous versus oral iron for the treatment on anaemia in IBD, five eligible studies, including 694 IBD patients, were identified []. IV iron demonstrated a higher efficacy in achieving a haemoglobin rise of ≥2.0 g/dL as compared to oral iron (OR: 1.57, 95% CI: 1.13, 2.18). Treatment discontinuation rates, due to adverse events or intolerance, were lower in the IV iron groups (OR: 0.27, 95% CI: 0.13, 0.59). Similarly, the occurrence of gastrointestinal adverse events was consistently lower in the IV iron groups. On the contrary, serious adverse events (SAEs) were more frequently reported among patients receiving IV iron preparations (OR: 4.57, 95% CI: 1.11, 18.8); however, the majority of the reported SAEs were judged as unrelated or unlikely to be related to the study medication. The recent European Crohn's and Colitis Organization (ECCO) guidelines [] conclude that ‘‘IV iron is more effective, shows a faster response, and is better tolerated than oral iron’’ and state that ‘‘IV iron should be considered as first line treatment in patients with clinically active IBD, with previous intolerance to oral iron, with haemoglobin below 100 g/L, and in patients who need erythropoiesis-stimulating agents; while oral iron may be used in patients with mild anaemia, whose disease is clinically inactive, and who have not been previously intolerant to oral iron []. The estimation of iron need is usually based on baseline haemoglobin and body weight (Table 5) [].

Anaemia seems to recur frequently and fast after intravenous iron therapy []. After successful treatment of iron deficiency anaemia with intravenous iron, re-treatment with intravenous iron should be initiated as soon as serum ferritin drops below 100 μg/L or haemoglobin below 12 or 13 g/dL according to gender [ECCO Anaemia Statement 3E].

Lately, there is interest in specific carbohydrate, paleolithic, gluten-free, low FODMAP, ω-3 PUFA enriched and other diets in active IBD. However RCT data regarding the effects of experimental diets on intestinal inflammation or on inducing remission are still lacking at this time. An adequately powered RCT of fructo-oligosaccharides (FOS) showed no clinical benefit in patients with active CD []. Therefore, no “oral IBD diet” can be generally recommended to promote remission in IBD patients with active disease. This recommendation does not prelude the needs of all IBD patients to receive an individual (nutritional) approach based on their specific personal situation, preferably with the active input of a dedicated dietician or nutritionist as part of the multidisciplinary approach. It is important that each IBD patient with active disease should undergo malnutrition screening and diet counselling in the case of malnutrition. It is recorded that approximately 75% of hospitalised CD patients suffer from malnutrition and 33% have a BMI <20 kg/m² []. Screening for nutritional deficiencies in chronic disease patients is warranted.

Enteral nutrition (EN), as an exclusive form of nutrition (EEN), has generated interest over 30 years as a treatment modality for active IBD since it is hypothesized to promote mucosal healing in the gastrointestinal tract by altering favourably the intestinal microbiota, reducing intestinal permeability, enhancing barrier defence and adaptation, and promoting a reduction of pro-inflammatory cytokines. In an open-label-trial in 37 CD children it was demonstrated that mucosa healing was significantly higher in the polymeric (74%; 95% CI 51%–89%) than the corticosteroid group (33%; 95% CI 16%–57%, P<0.05) []. Overall, studies were unable to show differences in remission rates between polymeric EN and elemental EN [, ]. EN in a supplemental form as partial enteral nutrition (PEN) therapy induced remission in 47 children and young adults [], whereas this effect was not found in a former RCT in 50 CD children []. Due to strong concerns over corticosteroid use and aiming for optimal growth in children, EN is often first-line therapy for paediatric patients with active CD []. Although EEN as primary therapy in adults with CD has also repeatedly been considered to be effective the data are not robust. Opposite results have appeared regarding the amount and nature of fat in the enteral formulae and on the question of polymeric versus elemental EN in RCTs of adults with active CD [, , ]. Meta-analyses do not support the use of EN as primary treatment for acute exacerbations of CD in adults [, ]. Patchy clinical conviction and the data, which appear better than might be expected with placebo, ensure continuing controversy over its role in adults.

In the case of extraordinary amount of faecal production, diarrhoea or increased/high output stoma (HOS), a systematic diagnostic approach is advised in which screening for clostridium, antibiotic associated diarrhoea, pouchitis in the case of IPAA, bile acid diarrhoea/steatorrhoea after distal ileal resection, (distal) colonic inflammation, lactase deficiency in the case of proximal small intestinal inflammation, and coeliac disease should be incorporated. Depending on the underlying cause of diarrhoea in IBD, medication can be considered as well as a supportive diet regime in some cases (eg lactose restricted diet).

Ongoing and severe diarrhoea or HOS can result in intestinal insufficiency [] with malabsorption, unintentional weight loss, malnutrition, nutritional deficiencies and/or dehydration. Malabsorption is an important contributing factor to malnutrition in IBD []. The retrospective study of Baker in 687 stoma patients [], showed that early high output (within 3 weeks) from an ileostomy is common and although 49% resolved spontaneously, 51% needed ongoing medical treatment, usually because of a short small-bowel remnant. 71% patients were treated with oral hypotonic fluid restriction, glucose-saline solution and anti-diarrhoeal medication to wean from parenteral infusions and 8% had to continue parenteral or subcutaneous saline in home-setting. Satisfactory home management with oral fluid restriction and monitoring of urine sodium content was demonstrated more than 35 years ago []. In a study in 13 adult (ileal) HOS patients, oral rehydration solutions containing rice maltodextrins (R-ORS) supplementation improved the sodium and potassium balance. The association of increased body weight with decreased serum renin concentrations suggests that a positive water balance also occurred []. In another study, 3 different saline and/or glucose solutions were tested in 6 patients with jejunostomies. Based on this small group, a sipped glucose electrolyte solution seemed to be the optimal mode of sodium replacement in patients with HOS []. No RCTs are available on nutritional treatment of IBD related diarrhoea or HOS. Only case studies on treatment of Crohn with HOS have been published, which show successful treatment with restriction of hypotonic fluids, sodium enriched diets, fully enteral nutrition and/or parenteral sodium-containing infusions.

Some patients with CD develop clinically significant intestinal strictures. Depending on their severity (degree of obstruction) and site, nutritional support may become necessary while the effects of treatment are awaited. Such treatment may be medical (with drugs) where the narrowing is mainly the result of inflammation, or mechanical (by balloon dilatation or surgery) when there is fibrotic scarring. In patients with radiologically identified but asymptomatic stenosis of the intestine it is conventional to recommend a modified diet which is low in insoluble fibre, but there are no robust data to support this apparently logical approach. When symptoms are present it may be necessary to adapt the diet to one of soft consistency, perhaps predominantly of nutritious fluids.

Intestinal fibrosis is a common feature of CD and may appear as a stricture, stenosis, or intestinal obstruction. Stenosing CD leads to a significantly impaired quality of life in affected patients and constitutes a challenging treatment situation. Different treatment approaches with potentially harmful side effects are frequently used: medical options (drugs) where the narrowing is mainly the result of inflammation, endoscopic (by balloon dilatation) or surgical approaches when there is fibrotic scarring. Depending on their severity (degree of obstruction) and site, nutritional support may become necessary while the effects of treatment are awaited at least in case of (risk of) malnutrition.

A recent Chinese prospective observational study in 59 adult CD patients with inflammatory bowel strictures showed that 12-weeks exclusive enteral nutrition (EEN) can effectively relieve inflammatory bowel strictures; (81.4%) achieved symptomatic remission, 35 patients (53.8%) achieved radiologic remission, and 42 patients (64.6%) achieved clinical remission []. A small study of 7 patients showed no clinical effect of TPN on colonic strictures []. No RCTs are available on nutritional management in IBD strictures. Some case studies report on occasional effectiveness of TPN or semi-elementary enteral nutrition.

Although it is common practice to recommend a modified diet with adapted consistency perhaps predominantly of nutritious fluids, at least in patients with radiologically identified stenosis of the (proximal) intestine and obstructive symptoms, or to feed distally by enteral nutrition whenever this is possible, there are no robust data to support these apparently logical approaches.

Osteoporosis (low bone mineral density BMD) and fractures are frequently encountered in patients with CD. The prevalence of osteoporosis in paediatric patients with IBD is approximately the same as in adult patients. Osteoporosis may already be present before steroid treatment []. In order to prevent fractures, treatment with bone protecting drugs appears warranted early in the course of bone disease when bone loss is not yet prominent. Significant risk factors for low BMD studied in adult IBD populations (n = 116 and n = 205) prove to be low serum vitamin D, male gender, Asian ethnicity, CD, low BMI and corticosteroid use, whereas no consensus on role of age, or age at diagnosis was found [, ]. In children and adolescents with IBD risk factors associated with low BMD are cumulative corticosteroid dose, height-for-age Z-score, and BMI Z-score [].

It should however be remembered also that prednisone treatment in CD can stimulate food intake, promoting an overall positive energy balance despite large faecal nutrient losses [].

There is no overall consensus on the vitamin D status and necessary actions in children and adolescents with IBD. In Veit's study there is no difference in mean serum 25(OH)D concentration between children and adolescents with IBD and controls (n = 58 child vs n = 116 HC) []. Vitamin D deficiency is common (55%) among adult patients with active UC, particularly those requiring corticosteroids (n = 34) []. Vitamin D deficiency should be treated since low plasma 25(OH)D is associated with an increased risk of surgery and hospitalizations in both CD and UC, and normalization of 25(OH)D status is associated with a reduction in the risk of CD-related surgery (n = 3217 adults with IBD) []. Next, a higher plasma 25(OH)D is associated with reduced risk of Clostridium difficile infection in patients with IBD (n = 3188 adults with IBD) []. Vitamin D supplementation seemed effective in increasing serum 25(OH)D levels in 83 children with quiescent CD [].

A RCT of 132 osteopenic CD patients, showed improved BMD at lumbar spine after 2 years of once weekly treatment course with risedronate 35 mg, concomitant with calcium and vitamin D supplementation []. An earlier RCT showed no significant benefit of calcium supplementation (1 g/day) alone on the BMD at 1 year in corticosteroid-using IBD patients with osteoporosis [].

Evaluation for vitamin D deficiency is recommended in IBD, and ensuring always an adequate supply of calcium and vitamin D, especially in steroid-treated IBD patients. Limitation of corticosteroid use helps to prevent low BMD.

The common causes of bile acid malabsorption are ileal resection and inflammation of the terminal ileum, common in CD. Decreased reabsorption of conjugated gall bile acids leads to excess transmission to the colon, where deconjugation by bacteria occurs. Osmotic diarrhoea and (in severe bile acid malabsorption) fat malabsorption might be a consequence []. If mild, bile acid diarrhoea can be controlled by a sequestrant such as cholestyramine [, ]. In a double-blind cross-over study in 14 CD patients who had undergone ileal resection, no negative effect of colestyramine treatment on jejunal fat absorption was reported. In severe cases of bile acid malabsorption however, steatorrhoea may worsen as a result of colestyramine treatment [].

Enteric (secondary) hyperoxaluria (with increased risk of kidney stones) occurs in severe small bowel CD associated with fat malabsorption and a consecutive elevation of intestinal oxalate absorption. Enteric hyperoxaluria may occur after ileal resection. Presence of the colon is an important factor, as oxalate remains available for colonic absorption because of concomitant fat malabsorption and its binding of calcium []. Urinary oxalate excretion correlates with fat excretion, as was shown in one study in CD patients undergoing intestinal resection. Increasing the dietary fat intake in these patients further increased urinary oxalate excretion []. Significantly lower mean values of urinary oxalate excretion were found in paediatric than in adult Crohn's patients []. A reason for this may be the shorter history of CD, which usually also implies fewer bowel resections. This implies that a diet low in fat and oxalate and high in calcium should be recommended in patients with hyperoxaluria. Restriction of dietary oxalate (teas and fruits mainly) seems warranted only in those with recurring urinary tract stones.

The systematic enquiry revealed insufficient evidence to make firm recommendations for exclusion diets as induction therapy. Exclusion diets have been described to alleviate symptoms [], but only few studies reports induction of remission [, ]. In the open label study by Sigall-Boneh et al., 47 paediatric and adult CD patients received polymeric formula feed (50% of caloric intake) combined with an exclusion diet (no gluten, dairy products, gluten-free baked goods and breads, animal fat, processed meats, products containing emulsifiers, canned goods, and no packaged products). After 6 weeks, remission was obtained in 70% of children and 69% of adults []. Another uncontrolled study in only 6 paediatric patients with moderate-severe CD, using an elimination diet (free of dairy products, certain grains and carrageenan containing foods) together with nutraceuticals (consisting of fish peptides, bovine colostrum, boswellia serrata, curcumin and a multivitamin) as well as Lactobacillus GG, and also growth hormone (administered daily) showed induction of remission in all patients [].

In a randomised controlled trial, longer maintenance of remission (after successful induction of remission using elemental formula) was seen in patients using a stepwise dietary introduction programme excluding foods that worsened symptoms, compared to patients receiving corticosteroids on a tapering schedule while eating a normal diet []. Similar results on maintenance of remission were reported in an open label study by the same group using a personal food exclusion diet []. Another study reported maintenance of clinical remission using a IgG4 guided exclusion diet in adult CD patients [].

Exclusion diets are labour-intensive for staff, and complex, challenging and often unpleasant for patients. The systematic enquiry revealed no evidence that exclusion diets are hazardous when applied under medical supervision. Evidence was not forthcoming to indicate that they contribute to nutritional deficiencies. Nonetheless it is good practice to monitor carefully for deficiencies that might be predicted from any particular set of exclusions.

Two clinical trials in paediatric UC patients show a moderate effect of rectal enemas containing Lactobacillus reuteri in mild distal colitis [] and of an oral preparation of VSL#3 in active colitis []. There are no specific data confirming harm, but lack of efficacy and the possible enhanced risks of and from bacteraemia in acute severe colitis lead the panel to advise against their use.

The systematic enquiry indicated that probiotics were, in general, ineffective in active CD. Not a single RCT has been performed using probiotics as induction treatment in paediatric CD. As stated in the recent ECCO/ESPGHAN guidelines on paediatric CD, probiotics are also not recommended for maintenance of remission []. It is possible that probiotics other than those studied or optimised doses and periods of treatment might have more useful effects, but the panel recommended that they should not be used. There are some positive data in respect of the use of Lactobacillus GG in maintenance in children with CD [].

The decision on the optimal route of artificial nutrition in IBD can be complex and involve several aspects, including the ability of the patient to eat, the absorptive capacity of the GI tract, the nutritional status of the patient, and the therapeutic goals (supportive care, treatment of malnutrition, induction of remission, maintenance of remission). The decision will also be influenced by the type of formula used in prior studies, and the dietary modulation of the intestinal immune response in IBD and its potential clinical implications.

Oral Nutrition Supplements (ONS) are the first step but generally are but a minor supportive therapy used in addition to normal food. By using ONS, a supplementary intake of up to 600 kcal/day can be achieved without compromising normal food intake in adults. Enteral feeding using formulae or liquids should always take preference over parenteral feeding, unless it is completely contraindicated. If oral feeding is not possible, feeding the patient through a nasogastric or nasoenteric tube should be considered.

Enteral nutrition should be considered in patients with a functional gastrointestinal tract but who are unable to swallow safely [, ]. In situations when the gut cannot absorb all nutritional needs, enteral nutrition should nonetheless be attempted with supplementary PN [, , ].

PN is indicated when there is an obstructed bowel where there is no possibility of placement of a feeding tube beyond the obstruction or where this has failed. It is required in patients with short bowel resulting in severe malabsorption of nutrients and/or fluid and electrolyte loss which cannot be managed enterally. PN is also indicated in surgical cases as above, and in any patient who is intolerant of enteral nutrition or in whom nutrition cannot be maintained by the enteral route []. However, it must be recognized that these patients in need of PN are those with the most complicated disease [].

There are strong clinical impressions supported by trials deemed to be of poor quality that primary nutritional therapy is effective in the induction of remission and that the remission rates are reproducibly better than might be expected from a placebo response. It is therefore recommended that primary nutritional therapy in the form of exclusive enteral nutrition (EEN) is considered in all patients with acute active CD and that this is a first choice in patients at high risk from alternative therapy such as steroids. Old meta-analyses demonstrated that corticosteroids are better than EEN in induction of remission in adults. The argument in favour of EEN is stronger in paediatric practice and will normally be the first choice in many centres. Firstly, this is because of the deleterious effects of undernutrition on growth []. Secondly, since growth is so essential in children, this increases the possibility of avoiding the use of steroids or delaying their introduction [] which is of paramount importance. Third, and most importantly, is the observed effect on induction of remission in paediatric studies demonstrating similar efficacy of steroids and EEN [], and that in some settings (i.e. concomitant immuno-modulatory treatment) EEN might even be superior to corticosteroids in children []. However, these studies suffer from major methodological limitations including lack of proper randomization and retrospective analysis. Furthermore, most of the data relate to mild to moderate disease activity.

Recommendations in children are made only for EEN as limited data suggest that partial enteral nutrition may be less effective than exclusive enteral nutrition [], though one RCT showed similar efficacy [].

The data are weaker for adult practice [], and most centres will continue to use steroids (or biologicals) as first-line therapy unless these agents are actively contra-indicated. However patient and disease characteristics also contribute to therapeutic management decisions and these may make enteral nutritional therapy a first-line option also in selected cases of adults with acute CD [].

EN is preferred, because PN has not been shown to offer any advantage in CD, and should be used only to improve nutritional status for surgery and when other modes of nutrition are not possible [].

There are few reliable data on special steps or complications peculiar to patients with IBD. Reference can be made to general guidelines for nutrition support in severely malnourished patients, in respect of both EN and PN. Some features specific to IBD can nonetheless be summarised.

Tube feeding can be safely delivered by nasogastric tube, or percutaneous endoscopic gastrostomy [, , ]. Continuous tube feeding administered via an enteral feeding pump and increased slowly to the full prescribed volume appears to have lower complication rates than bolus delivery [, , , ]. The most frequent complications of EN are mechanical (tube-related), then metabolic and infectious, but these are not notably different from those seen in other chronic conditions [, ].

Few patients with UC will need artificial feeding other than during the most severe exacerbations and in the peri-operative phase. Enteral nutrition is most appropriate and associated with significantly fewer complications than parenteral nutrition in acute colitis. Bowel rest through intravenous nutrition does not alter the outcome, but nonetheless, there are no specific contraindications for the use of parenteral nutrition in UC.

In CD nutritional support is more often needed. Specific micronutrient deficiency states are relatively common in CD; these should be sought (perhaps annually) and corrected as appropriate – a need for supplementary iron (oral or intravenous) and for parenteral vitamin B12 being the most common.

There is no specific contraindication to the use of parenteral nutrition in patients with CD in comparison to other diseases, and a central or peripheral route may be selected according to its expected duration. There are not enough data to dictate the use of specific substrates in the composition of PN in CD. PN must however be adjusted to fulfil the needs of the individual patient. This will reflect the extent of malabsorption, and enteric losses, and will influence the prescription of energy and amino acids, and especially of water, electrolytes and minerals. Each PN cycle (usually nocturnal) should be complete and adjusted according to progress (eg through the number of cycles per week). PN, especially at home, should be viewed as complementary non-exclusive nutrition, which can be tapered to a minimal level when body composition has been sufficiently restored. The most frequent complications of PN in IBD are infectious (catheter sepsis), metabolic and mechanical. Specific attention should be paid to electrolyte supplementation (especially sodium and magnesium) in short bowel patients. Again, these risks and precautions are not notably different from those seen in other chronic conditions.

Several studies have compared the efficacies of different types (elemental, semi-elemental, oligomeric or polymeric diets) of enteral formulae in the management of active CD. A Cochrane meta-analysis of ten trials showed no statistically significant difference between patients treated with elemental (n = 188), and non-elemental diet (semi-elemental or polymeric diet; n = 146) []. The protein composition did not appear to influence the therapeutic potential of EN. The present systematic enquiry reveals insufficient evidence to make firm recommendations [, ]. It is therefore advised that standard feeds are employed if primary nutritional therapy is being employed. There are hypothetical advantages from some amended formulations.

Comparing one form of enteral nutrition to another has not shown any difference in effectiveness for treating active CD, but a non-significant trend favouring low fat formulations has emerged [, , ]. Some centres may therefore wish to consider the use of feeds with lower fat content.

The use of feeds supplemented with growth factors, ones with lower levels of emulsifying data, or oligomeric feeds, as alternatives to standard feeds, is not supported by reliable data [, , ]. Equally there is no evidence that any of these alternatives is inferior to the use of standard polymeric feeds [, ].

There are not enough data to dictate the use of specific substrates in the composition of PN in CD. PN must however be adjusted to fulfil the needs of the individual patient. This will reflect the extent of malabsorption, and enteric losses, and will influence the prescription of energy and amino acids, and especially of water, electrolytes and minerals. Each PN cycle (usually nocturnal) should be complete and adjusted according to progress (eg through the number of cycles per week). PN, especially at home, should be viewed as complementary non-exclusive nutrition, which can be tapered to a minimal level when body composition has been sufficiently restored [, , ]. The most frequent complications of PN in IBD are infectious (catheter sepsis), metabolic and mechanical []. Specific attention should be paid to electrolyte supplementation (especially sodium and magnesium) in short bowel patients [, ]. Again, these risks and precautions are not notably different from those seen in other chronic conditions.

Patients with IBD are at increased risk of venous thromboembolism. Thrombosis is a specific feature of IBD that can be involved in both the occurrence of thromboembolic events and the pathogenesis of the disease itself [, ]. The precise aetiology for the higher rates of thromboembolism in IBD and the specific association is as yet unknown, but multiple acquired and inherited factors are implicated. The impact of inflammation on coagulation has been confirmed by several experimental studies showing that inflammatory mechanisms shift the haemostatic balance to favour the activation of coagulation which, in turn, can also sustain inflammation promoting a vicious circle between chronic inflammation and thrombosis. Although there are insufficient data to mandate routine anticoagulation, this should be considered in all IBD patients and especially those on PN, with every effort made to avoid dehydration [, , , , ].

Patients with CD are prone to fistulae formation between 2 intestinal sites or from intestine to another organ (especially skin, bladder and vagina). Most occur post-operatively. It is demonstrated that in surgical patients, early nutritional support, independently of the route of administration, decreases the occurrence and severity of fistulae [, , ]. Malnutrition with BMI <20 appears as an independent risk factor that should be confirmed in further studies [].

Treatment of intestinal fistulae is usually complex, depending on the location, scale and the nature of the symptoms, and warrants the input of a multidisciplinary team including gastroenterologist, surgeon and dietician []. Treatment will often need to be surgical but some patients clearly benefit from drug treatment with immunomodulators or/and biologics [, ]. Once a fistula is mature and there is no longer any possibility of a free communication with the peritoneal space, there ceases to be any contraindication to enteral nutrition. Indeed in the patient with a distal (low ileal or colonic) fistula it may be possible to provide all necessary nutritional support via the enteral route [, , ]. In the patient with a proximal fistula and/or a very high output it may be preferable to manage the situation with a rested gut and full PN [, ], but even then the psychological benefit of eating may warrant its inclusion in the nutritional regimen despite minimal expectations of useful nutrient absorption []. Surgical correction is more likely to be successful if nutritional status has been optimised pre-operatively [].

Refeeding syndrome should not be a problem in the well-managed patient with IBD but nonetheless it is not unusual to encounter patients in whom nutritional deprivation has extended over many days and in whom this hot issue is pertinent. Standard precautions and interventions are mandatory in these high-risk patients particularly in respect of phosphate and thiamine [, , ].

The systematic enquiry demonstrated evidence in favour of the use of probiotics in induction of remission and in maintenance of UC – see elsewhere in this document.

Despite early indications that omega-3 fatty acid supplementation contributed beneficially in induction and maintenance the systematic enquiry documented an absence of effect from a diet supplemented by omega-3 fats in patients with UC in the maintenance of remission [, , , , , ]. This is therefore not advised.

The above data were obtained in adults. It appears reasonable and safe to extrapolate the conclusions and suggested actions on omega-3 fats into paediatric practice.