Nutrition and adult inflammatory bowel disease
Correspondence to: Dr J. Goh, Selly Oak Hospital, Radderbarn Road, Selly Oak, Birmingham, B29 6JD, UK. E-mail: Jason.Goh@uhb.nhs.uk
Major advances in the understanding of the aetio-pathogenesis and genetics of inflammatory bowel disease have been accompanied by an escalation in the sophistication of immunomodulatory inflammatory bowel disease therapeutics. However, the basic ‘triple’ therapy (5-aminosalicylates, corticosteroids, azathioprine) and nutrition have maintained their central role in the management of patients with inflammatory bowel disease over recent decades.
This review provides an overview of the supportive and therapeutic perspectives of nutrition in adult inflammatory bowel disease.
The objective of supportive nutrition is to correct malnutrition in terms of calorie intake or specific macro- or micronutrients. Of particular clinical relevance is deficiency in calcium, vitamin D, folate, vitamin B12 and zinc.
There is justifiably a growing sense of unease amongst clinicians and patients with regard to the long-term use of corticosteroids in inflammatory bowel disease. This, rather than arguments about efficacy, should be the catalyst for revisiting the use of enteral nutrition as primary treatment in Crohn's disease.
Treatment failure is usually related to a failure to comply with enteral nutrition. Potential factors that militate against successful completion of enteral nutrition are feed palatability, inability to stay on a solid-free diet for weeks, social inconvenience and transient feed-related adverse reactions. Actions that can be taken to improve treatment outcome include the provision of good support from dietitians and clinicians for the duration of treatment and the subsequent ‘weaning’ period. There is evidence to support a gradual return to a normal diet through exclusion–re-introduction or other dietary regimen following the completion of enteral nutrition to increase remission rates. We also review the evidence for emerging therapies, such as glutamine, growth factors and short-chain fatty acids.
The future may see the evolution of enteral nutrition into an important therapeutic strategy, and the design of a ‘Crohn's disease-specific formulation' that is individually tailored, acceptable to patients, cost-effective, free from adverse side-effects and combines enteral nutrition with novel pre- and pro-biotics and other factors.
Recent decades have witnessed significant progress in the understanding of the aetio-pathogenesis, genetics and therapeutics of inflammatory bowel disease. The mainstay of medical treatment, however, has remained relatively unchanged in the form of anti-inflammatory (mesalazine) and immunosuppressive (corticosteroids and thiopurine) drugs.1, 2 Nutrition plays a pivotal role in the clinical care of all patients with inflammatory bowel disease. In broad terms, nutritional therapy can be considered as supportive or primary treatment. Supportive therapy aims to correct malnutrition and macronutrient deficiencies and reverse their metabolic/pathological consequences, in addition to providing advice on specific dietary regimes. This should be considered in all patients with inflammatory bowel disease. Nutrition as primary treatment, on the other hand, has a more restricted repertoire and can benefit specific subgroups of patients. The use of enteral nutrition as a primary treatment in the paediatric population with Crohn's disease is widely practised and has been considered elsewhere.3–5 In this article, we review the role of nutrition as supportive and primary therapy in specific clinical scenarios in adult inflammatory bowel disease.
The pattern and severity of malnutrition are dependent on the duration, activity and geographical extent of the disease. Fundamental differences in the pattern of malnutrition exist between Crohn's disease and ulcerative colitis. Disease affecting the small bowel tends to lead to a greater incidence of protein–energy malnutrition and specific nutrient malnutrition than does colonic disease. In addition, patients with Crohn's disease generally develop malnutrition over a long period of time, whereas patients with ulcerative colitis tend to be relatively well nourished whilst in remission, but can develop precipitous nutritional deficiency during a hospital admission for a severe acute relapse.6
Factors contributing to malnutrition in inflammatory bowel disease include reduced oral intake, malabsorption, increased nutrient losses from the gut, drug–nutrient interaction and increased requirements. The reported prevalence of protein–energy malnutrition in inflammatory bowel disease is in the range 20–85%.6–10 Hypoalbuminaemia is found in 25–80% and 25–50% of hospitalized patients with Crohn's disease and ulcerative colitis, respectively.11 There is a lack of consensus as to whether the basal metabolic rate is increased in inflammatory bowel disease. Increased resting energy expenditure, increased lipid oxidation, reduced glucose oxidation and decreased diet-induced thermogenesis have been reported in patients with both active12 and inactive13 Crohn's disease. Increased resting energy expenditure has been observed in patients with ulcerative colitis and Crohn's disease by some investigators,14, 15 but not others.16
Specific nutrient deficiency
A wide array of vitamin and mineral deficiencies occurs in inflammatory bowel disease, with varying degrees of clinical significance. Of particular relevance to clinicians is the impact of nutrient deficiency on bone mineral density, carcinogenesis and thrombophilia associated with inflammatory bowel disease.
Osteopenia and osteoporosis are increasingly being recognized as leading extra-intestinal morbidities amongst patients with inflammatory bowel disease.17 Current data on the prevalence of osteoporosis in inflammatory bowel disease vary widely due to cohort heterogeneity and a lack of uniform diagnostic criteria and methodology.18–21 Using fracture as a meaningful end-point, Bernstein et al. showed a 40% increase in the risk of fractures amongst patients with inflammatory bowel disease compared with sex- and age-matched controls,22 whilst Vestergaard et al. reported a 2.5-fold increase in the risk of fracture amongst women with Crohn's disease, but not amongst men with Crohn's disease or patients with ulcerative colitis.23 Many factors, other than dietary calcium or vitamin D deficiency, contribute to the pathogenesis of osteoporosis in inflammatory bowel disease, such as lifetime steroid use, body mass index, disease activity and hormonal and genetic factors. However, this topic is considered in the current review as nutrition forms an integral part of the prophylaxis and treatment of osteoporosis in inflammatory bowel disease.
Historically, long-standing colitis has been considered as a risk factor for colitis-associated carcinogenesis (reviewed in Ahnen24 and Leonard-Jones25). Although there is no direct evidence to link any dietary factor and cancer risk, folate is by far the most promising putative anti-cancer nutrient. Epidemiological data have implicated an inverse relationship between dietary folate intake and sporadic colorectal cancer.26–28 In patients with inflammatory bowel disease, folate deficiency may result from dietary insufficiency, enhanced intestinal loss or, less frequently, from competitive inhibition with concomitant sulfasalazine therapy. Two case–control studies29, 30 and a retrospective analysis31 have shown a protective effect of folate supplementation against dysplasia and cancer in patients with long-standing ulcerative colitis. There are currently no placebo-controlled studies of folate, antioxidant vitamins or selenium supplementation in the prevention of colitis-associated cancer.
Another adverse effect of folate deficiency is hyperhomocysteinaemia and its associated prothrombotic state. It has been known for some time that the incidence of arterial and venous thrombo-embolic events is elevated in inflammatory bowel disease.32, 33 Hyperhomocysteinaemia, a known inducer of a hypercoagulable state, has been reported in 26.5% of patients with inflammatory bowel disease, compared with only 3.3% of controls.34 This has been confirmed by other investigators.35–37 The most striking feature from these studies is the identification of a low serum folate level (and, to a lesser extent, a low serum vitamin B12 level) as a strong independent risk factor for hyperhomocysteinaemia.34–37 Both folate and vitamin B12 are crucial co-factors in the homocysteine–methionine metabolic pathway.35
Other nutritional deficiencies that should be sought in appropriate clinical settings are iron, zinc and selenium. Zinc is crucial for wound healing, and its deficiency may be a factor to consider in patients with Crohn's disease presenting with persistent fistulas.38, 39 In addition, it is a co-factor for superoxide dismutase and may be protective against free radical-mediated cellular damage.
Supportive nutritional therapy
Clinicians should have a high index of suspicion for malnutrition in both in- and out-patients with inflammatory bowel disease. Nutritional assessment entails a combination of history-taking, clinical examination and laboratory tests. The body mass index, serial body weight, presence of oedema, serum albumin and iron studies provide a quick guide to the patient's nutritional status. A more formal assessment may be undertaken by the clinical nutritionist or clinician in the form of anthropometry and the use of scoring systems, such as the Nutrition Risk Score40 or Subjective Global Assessment.41
Specific nutrient deficiency should be sought where clinically indicated. Barak et al. have shown that patients hospitalized for active inflammatory bowel disease have an average stress factor of 1.05–1.10, in agreement with previous studies which showed little increase in energy requirement beyond that predicted by the Harris–Benedict equation for patients with active inflammatory bowel disease without abscess or sepsis.42
Enteral and parenteral supportive nutrition
The primary objective of supportive nutrition is to improve the nutritional status of a malnourished patient, and, unless absolutely contraindicated, the enteral route is preferred. Increasing oral intake with nutritional supplements, food fortification and a high-protein, high-energy diet may be considered initially to improve nutritional status. Nasogastric pump feeding may be instituted if the oral ingestion route is not possible. This may be due to patient preference, severe anorexia and nausea, oral disease or simply to ensure adequate calorie intake.
For the purpose of supportive nutrition, there is little to choose between the various commercial preparations, although elemental feeds have the added advantage of therapeutic efficacy in active Crohn's disease (discussed below). The choice depends on palatability, cost considerations, patient tolerance, osmolality and specific calorie requirements. The provision of calories from carbohydrate, protein and fats in inflammatory bowel disease patients does not differ from that of other patient groups requiring nutritional support. However, severely malnourished patients in an obvious catabolic state will require the appropriate adjustment factor when calculating their calorie needs, and particular attention must be paid to the correction of electrolyte abnormalities prior to the start of feeding in order to minimize the metabolic entropy encountered in the re-feeding syndrome. Some patients will also benefit from nutritional support pre-operatively. Again, this should be delivered via the enteral route if at all possible.
The use of total parenteral nutrition should be restricted to patients in whom enteral nutrition is contraindicated or, as discussed below, as therapy in specific clinical circumstances. Total parenteral nutrition adds significantly to the cost and length of hospital stay, and is associated with many complications related to the use of central venous catheterization, such as sepsis, venous thrombosis, pneumothorax and lymphatic duct damage. Animal data also suggest an increased rate of bacterial translocation with a prolonged lack of bowel use.43–45 In inflammatory bowel disease, total parenteral nutrition is used to deliver nutritional support on a short-term basis in patients with bowel obstruction or perforation, toxic megacolon or pre-operatively, and on a long-term basis in patients with short bowel syndrome. The use of total parenteral nutrition pre-operatively has been shown to improve serum albumin and body weight in patients with Crohn's disease,46 but its impact on post-operative morbidity and mortality remains unclear. It has been suggested that pre-operative total parenteral nutrition may decrease post-operative complications in severely malnourished patients with inflammatory bowel disease only when given for at least 5 days.6 Pre-operative total parenteral nutrition in Crohn's disease has been investigated primarily in retrospective studies. These have shown fewer post-operative complications, an improved clinical course and a decrease length of bowel requiring resection, but at the expense of a longer hospitalization.47
Home parenteral nutrition
In inflammatory bowel disease, the most common indication for home parenteral nutrition is chronic intestinal failure secondary to short bowel syndrome, usually, but not always, as a consequence of multiple small bowel resections for Crohn's disease.48–50 In a retrospective analysis of 41 patients with Crohn's disease, referred for home parenteral nutrition to a specialized intestinal failure unit in the UK, the causes of intestinal failure were multiple small bowel resections in 83% and extensive primary Crohn's disease in 17%.48 Short bowel syndrome usually occurs when there is less than 200 cm of viable jejunum left, which is insufficient to sustain adequate absorptive function of fluids, electrolytes and nutrients. Home parenteral nutrition is usually necessary when the length of the remaining small bowel is less than 100 cm.51, 52
Starting a patient on home parenteral nutrition is a major and costly undertaking. It requires full education of the patient and the relevant carers, who should be supported by a multi-disciplinary team comprising a physician, nurse specialist, clinical nutritionist and pharmacist, with input from the clinical microbiologist, interventional anaesthetist and radiologist for the joint management of catheter-related complications. Alternative models of care may include over-the-telephone consultation and domiciliary visits.53 There are many ways to provide the calories and these should be individually tailored. We use a 40%/60% split between lipids and carbohydrate, but some authorities advocate a higher contribution of calories from lipids.
A high vigilance for electrolyte, vitamin and trace element deficiency must be maintained, and may be prevented by elective administration of intravenous multi-vitamin preparations. Rare metabolic complications of home parenteral nutrition in patients with Crohn's disease, such as manganese intoxication,54 copper deficiency,55 Wernicke's encephalopathy56 and dry beri beri,56, 57 have been reported.
Although home parenteral nutrition delivers life-saving nutritional support to patients with short bowel syndrome, it is associated with significant morbidity and potentially life-threatening complications. Complications may be considered as catheter-related, metabolic, gastrointestinal, hepatic, renal and skeletal.58 In a retrospective series of 41 patients on home parenteral nutrition for Crohn's disease over an 11-year period (totalling 121 patient-years of home parenteral nutrition), 58.5% of patients had one or more home parenteral nutrition-related complications necessitating hospitalization.59 The reasons for hospital admission included catheter-related sepsis (in 79.2%), mechanical problems of the catheter (62.5%) and dehydration and/or electrolyte abnormality (20.8%). There were eight deaths in total and only 7% of the deaths were directly caused by catheter-related sepsis.
Although home parenteral nutrition was not associated with any significant increase in body weight or decrease in the frequency of further surgery for Crohn's disease, it was significantly associated with an increased quality of life, serum albumin and transferrin levels, and a reduction in requirements for oral steroids.58 A putative mechanism for catheter-related sepsis, other than that caused by Staphylococcus species, is bacterial translocation.60 An association between the length of remaining bowel and the frequency of catheter-related sepsis has been found. Specifically, a length less than 50 cm has been associated with a significantly higher frequency of sepsis caused by Enterobacter, Klebsiella, Pseudomonas and Proteus species.61 It is crucial to maintain and preserve venous access in patients on home parenteral nutrition as a loss of access for any reason is a major indication for referral of these patients for consideration for small bowel transplantation.50, 61
Other supportive nutritional regimes
General considerations. For patients with ulcerative colitis or Crohn's disease in remission, a healthy balanced diet with few restrictions should be encouraged. It is important for those with stricturing Crohn's disease to adhere to a low-residue diet to prevent obstructive episodes. During relapses characterized by diarrhoea, it is also prudent for patients to temporarily reduce their oral intake of dietary fibre, although this is not supported by controlled trials. A low-fibre diet may also be beneficial in the subset of patients with co-existing inflammatory bowel disease and irritable bowel syndrome, in whom abdominal symptoms are attributable to bacterial fermentation of insoluble dietary fibre.62, 63 There are currently no prospective studies of the effect of restriction of the amount or type of dietary fibre on abdominal symptoms in subjects with inflammatory bowel disease in remission.
It is the practice of some gastroenterologists to proscribe dairy products during flares of inflammatory bowel disease. Two studies have demonstrated an increased prevalence of lactose intolerance in patients with Crohn's disease.64, 65 In a study examining 260 patients with inflammatory bowel disease, Mishkin et al. found the prevalence of lactose malabsorption to be 40% in patients with Crohn's disease, compared to 29.2% in controls at low ethnic risk for lactose malabsorption and 13.3% in patients with ulcerative colitis.64 Furthermore, patients with Crohn's disease confined to the terminal ileum are more likely to have lactose malabsorption than are those with ulcerative colitis or Crohn's colitis. This concurred with a German study, which found a prevalence of lactose malabsorption of 46.9% amongst patients with Crohn's disease, with a high frequency (83.3%) in those with active disease.65 It is noteworthy that lactose malabsorption in Crohn's disease may be independent of duodenal lactase activity, and may be a result of small bowel bacterial overgrowth and increased transit time.65, 66
A case–control study has demonstrated that a polymeric enteral diet rich in soya milk and free from lactose significantly improves body mass composition in Crohn's disease patients who are lactose intolerant.67
We propose that, irrespective of ethnicity, lactose malabsorption should be sought in patients with Crohn's disease involving the small bowel who present with persistent diarrhoea. This can be easily detected by hydrogen breath test which has a high sensitivity (70.4%) and specificity (95.6%).65 An appropriate low-lactose diet may then be advised.68 An alternative approach, such as soya milk fortified with calcium or lactase enzymes, may be considered. There is currently no evidence to support the use of a low-lactose diet for patients with ulcerative colitis or in whom Crohn's disease is confined to the large bowel.64, 69
Dietary calcium. All patients with inflammatory bowel disease should be advised to consume a diet rich in calcium as a simple, effective means of prophylaxis against osteopenia and osteoporosis. A total daily intake of 1.5 g of dietary calcium should be encouraged. If there is inadequate dietary calcium intake, oral supplementation with 500–1000 mg calcium tablets may be prescribed.70, 71 In addition, vitamin D deficiency should be sought and treated as appropriate. A suitable preparation is Calcichew-D3 Forte (Shire Pharmaceuticals, Basingstoke, UK), which provides 1000 mg calcium and 800 i.u. of vitamin D.
Few studies have examined the effectiveness of dietary supplementation on bone loss retardation or the treatment of established osteoporosis in inflammatory bowel disease. There is evidence that, in patients with inflammatory bowel disease and established osteoporosis, calcium supplementation alone may be ineffective in improving bone mineral density.72, 73 In a prospective, placebo-controlled study of oral calcium and vitamin D supplementation in steroid-using patients with inflammatory bowel disease and osteoporosis, Bernstein et al. failed to demonstrate a beneficial effect on bone mineral density after 1 year of treatment.72 It is noteworthy that the dietary calcium intake of the study subjects approximated that of the recommended daily intake at the outset of the trial, and a bisphosphonate may be more appropriate in this setting.
Likewise, in another study, involving 33 patients with Crohn's disease, there was no change in the bone mineral density after 1 year of oral calcium/vitamin D supplementation.73 However, the addition of oral fluoride to calcium/vitamin D was associated with a significant improvement in bone mineral density in this study. On the other hand, it has been shown that supplementation with oral vitamin D prevents bone loss in patients with Crohn's disease at high risk of developing osteoporosis, even in those with normal serum vitamin D levels.74
Dietary supplementation is an important aspect of the overall strategy for the prevention and treatment of bone loss in inflammatory diseases associated with a high prevalence of corticosteroid use and accelerated bone loss, as laid down in American and European guidelines.70, 71, 75–77
Dietary folate. It has been proposed that oral folate supplementation should be recommended for patients with inflammatory bowel disease as an anti-neoplastic28–31 and anti-thrombotic34–37 agent. Despite studies linking folate deficiency to increased cancer risk in inflammatory bowel disease,29–31 there are currently no prospective studies demonstrating a reduction in cancer risk in colitis following folate supplementation. Indirect evidence may be derived from a case–control study involving patients with ulcerative colitis, with pancolitis and a disease duration of greater than 8 years, which showed that folate supplementation was associated with a relative risk for neoplasia of 0.72.31 Furthermore, there was a dose–response relationship between the folate dose and the risk of carcinogenesis.
Folate deficiency is associated with DNA damage and is a putative mechanism in the pathogenesis of colitis-associated cancer.78–80 In a prospective, placebo-controlled study of folate supplementation in 20 patients with colonic adenoma, 5 mg of daily folate was associated with an increase in the extent of genomic DNA methylation and a decrease in the extent of p53 strand breaks, both bio-markers of epithelial dysplasia, at 6 and 12 months of the study.81 Intriguingly, these effects were also observed in the placebo group, but only at 12 months. Cravo et al. also demonstrated a reduction in DNA hypomethylation in the rectal mucosa of patients with colonic adenoma with folate supplementation, which returned to baseline values on switching over to placebo in a cross-over trial design.82
Although not commonly detected in ulcerative colitis, microsatellite instability has been proposed as a carcinogenic mechanism in colitis-associated cancer. Of 23 patients with long-standing ulcerative colitis studied by Cravo et al., the three (13%) patients found to be positive for microsatellite instability were noted to have 30–50% lower serum, blood and colonic folate concentrations and lower dietary folate intake in the preceding 5 years.83 These changes were reversible in the one patient given folate supplementation for 6 months.
Supplementation with oral folinic acid (15 mg/day) has also been shown to reduce cell proliferation/kinetics in the rectal mucosa of patients with ulcerative colitis.84
It is unclear how these molecular modifications tie in with the important issue of risk reduction of neoplastic transformation. No studies have been performed to examine the impact of folate supplementation on the risk and incidence of thrombo-embolic events. Nevertheless, folate deficiency should be sought and corrected where appropriate.
Other dietary considerations. Infectious diarrhoea is a known precipitant of relapses in inflammatory bowel disease.85, 86 Patients with inflammatory bowel disease may be advised to avoid eating foods that are associated with an increased risk of transmitting pathogens known to cause gastroenteritis. Specifically, uncooked meat or raw eggs are best avoided. Dietary precautions also need to be taken during travel to areas with poor standards of food handling practice.
Few studies have examined the effect of a modification of diet in patients from different ethnic backgrounds. A questionnaire study from Leicester, UK, indicated that the majority of Hindus with inflammatory bowel disease have altered their diet and are less likely to consume spices or starch.87 The study also identified a link between a high intake of refined carbohydrate and Crohn's disease in south Asians. As discussed above, it is reasonable to exclude dietary lactose in Asians and Oriental patients with Crohn's disease, but there is little evidence to advocate any other major changes to a ‘traditional’ diet.
Nutrition as primary therapy
Enteral nutrition in Crohn's disease
In 1984, it was reported that the substitution of the diet of patients with acute Crohn's ileitis with an elemental diet produced comparable remission rates to those obtained with corticosteroids;88 this paved the way for an alternative treatment approach to conventional pharmacological immunomodulation. Although enteral nutrition has an established role in the treatment of Crohn's disease in the paediatric population,3–5, 89 its use in adult patients has been a matter of considerable debate.
There are currently three published meta-analyses90–92 and a Cochrane Database Systematic Review93 of randomized controlled trials of enteral nutrition in Crohn's disease, in addition to a meta-analysis of trials involving children only.89 The critical issues are: (i) how does enteral nutrition compare with corticosteroids?; (ii) what are the differences between different formulations of enteral nutrition?; (iii) what are the benefits of enteral nutrition over corticosteroids?; (iv) how should enteral nutrition be delivered such that it induces and maintains remission?; and (v) how should the ideal ‘enteral nutrition cocktail’ be optimized?
In a systematic review of four published randomized controlled trials, with a total of 130 patients in the enteral nutrition arm and 123 receiving corticosteroids, the pooled odds ratio was 0.3 favouring steroid therapy.93 The number of patients needed to treat with steroid rather than enteral nutrition to achieve one remission was four. The conclusion that corticosteroids are superior to enteral nutrition in inducing remission was upheld following an analysis that included unpublished abstracts,92 and concurred with earlier published meta-analyses.90–92
There are several caveats to the application of this observation and to the dismissal of enteral nutrition on the grounds that it is less efficacious than corticosteroids. The conclusion was based on an intention-to-treat analysis, and patients who successfully completed enteral nutrition achieved comparable remission rates to those given corticosteroids. It is noteworthy that the overall remission rate with enteral nutrition was in the region of 60%, which is higher than the 20–30% response rate observed with placebo.9, 94, 95 A potential confounding factor in a meta-analysis is the heterogeneity of the formulations studied. A retrospective study from a single centre in the UK, using a specific elemental formulation in 113 patients with Crohn's disease over a 10-year period, cited a remission rate of 85% with enteral nutrition, which is well within the efficacy range of drug therapy.96 Furthermore, a comparison of efficacy alone between enteral nutrition and corticosteroids is insufficient, as the two treatment modalities possess entirely contrasting safety profiles. Enteral nutrition has no known long-term adverse effects, whereas the morbidity associated with steroids is multi-factorial.17, 97 Although there is a general consensus on the need to avoid the use of corticosteroids in children due to their effect on growth, there is an irrational sense of complacency about the use of corticosteroids in adults.
Dear et al. demonstrated that patients with Crohn's disease treated predominantly by diet or other steroid-sparing strategy, with a low lifetime corticosteroid dose, had a bone mineral density comparable with that of age-matched controls.98 Not surprisingly, the group of patients treated predominantly with corticosteroids had a significantly lower bone mineral density. Increasingly, clinicians will have to justify their prescription of corticosteroids for inflammatory bowel disease as the lay public learns about the complications associated with their long-term use.
Mechanism of action. On the issue of formulation, in a systematic review of nine trials comparing 170 patients treated with an elemental diet and 128 patients treated with non-elemental formulations, no significant difference was detected.93, 99 This may imply that the therapeutic effect of enteral nutrition is independent of the type of nitrogen source. Differences in the composition of long-chain triglycerides100 and polyunsaturated fatty acids101 are potential contributing factors influencing the efficacy of enteral nutrition.
A hypothesis for the mechanism of action of enteral nutrition in inducing remission in Crohn's disease is decreased antigenic presentation, leading to the alteration of the gut flora and modulation of the intestinal mucosal immunophenotype. The exclusion of microparticles, often found in an unrefined diet, may, in part, account for the efficacy of enteral nutrition.102 Repletion of nutrient deficiency is another potential mechanism. However, the exact mechanisms by which enteral nutrition induces remission in Crohn's disease are currently unclear.
Practical considerations. Adult patients with Crohn's disease who should be considered for enteral nutrition include: (i) patients in whom there is a potential for a high lifetime corticosteroid dose, including adolescents to patients in their thirties; (ii) patients considered to be at high risk for osteoporosis; (iii) patients who are steroid-refractory, steroid-dependent or steroid-intolerant; and (iv) patients who request alternative treatment. Having identified the patient who may benefit from enteral nutrition, the next step is to assess the patient in terms of his or her likelihood of compliance and the practicality of implementing the regime. This may be difficult, and studies examining factors that predict compliance and treatment success are eagerly awaited.
The main concerns in initiating enteral nutrition are adult patient compliance and dropout rate, an under-resourced clinical nutrition department and physician scepticism. Factors contributing to dropout include the palatability of feeds, length of time without solid food, social inconvenience and the transient side-effects associated with enteral nutrition (Table 1).103 Patients with active Crohn's disease should be offered a ‘balanced’ view, so far as is possible, with regard to the relative pros and cons of enteral nutrition vs. corticosteroids, and should be allowed to make their own choice. Identification of the ‘motivated’ patient is important to enhance compliance, although there are no published data to support this view. A patient who opts for enteral nutrition should be told that it may take longer than corticosteroids to induce remission, as this may reduce the expectation of a ‘quick fix’, which is often experienced when corticosteroid is commenced.
Table 1. Factors contributing to treatment failure
| Transient side-effects: headache, nausea, dizziness, diarrhoea|
| Level of motivation|
| Inability to cope with length of time without solid food|
| Social inconvenience|
| Stricturing disease or bulky inflammatory mass requiring surgery|
| Human resources issue (dietitians)|
| Lack of support (telephone-in facility)|
| Lack of robust protocols|
| Physician scepticism|
The choice of formulation is dependent on patient preference, palatability, osmolality, specific calorie requirement and centre protocol. Elemental feeds include E028 (Scientific Hospital Supplies, Liverpool, UK) and Vivonex (Norwich Eaton, Surrey, UK). Semi-elemental preparations, such as Peptamen (Clintec Nutrition, Chicago, IL, USA), Pepdite 2+ (Scientific Hospital Supplies, Liverpool, UK) and Perative (Abbott Laboratories Limited, Dublin 24, Ireland), are often more suitable for nasogastric tube feed.
With regard to the fat composition of enteral nutrition, a recent, prospective, randomized controlled trial suggested that the amount of medium-chain triglyceride may not influence the improvement rates of Crohn's disease when given in liquid form.104
A dietitian should ideally be present to assess and educate the patient on the same day, and should be available for telephone consultation for the duration of treatment. A decision regarding oral or nasogastric tube feed is taken jointly with the patient, the latter rarely being needed in adults. The dietitian's clinical report should be formally filed as part of the patient's medical record to ensure a multi-disciplinary input to the treatment and continuity of patient care. Guides for clinicians and dietitians in the implementation of enteral nutrition are available.99, 103
Improving treatment outcome. Clinicians need to be clear about the objective of treatment with enteral nutrition. Enteral nutrition may be used to: (i) induce remission and possibly maintain remission; (ii) induce remission as a ‘bridge’ to pharmacological immunosuppressive therapy; or (iii) to reduce corticosteroid dependence.
The observation that patients relapse quickly following the resumption of a normal diet,105, 106 and the relatively high relapse rates at 1 year of 65–100%,90–92,105 have prompted the search for strategies to maintain the remission achieved after 2–6 weeks of enteral nutrition. In a pilot study involving 20 patients, comparing an exclusion diet and an unrefined carbohydrate fibre-rich diet, Alun Jones et al. found that 70% of patients maintained on an exclusion diet remained in remission at 6 months, whereas 100% of the ‘normal’ diet group relapsed by 6 months.105 In a multi-centre trial, 78 patients with active Crohn's disease, who achieved remission with enteral nutrition, were randomized to receive corticosteroid or an exclusion and re-introduction diet.106 Relapse rates at 2 years were lower for the diet group (62% vs. 79%; diet vs. corticosteroid; P < 0.05). The median lengths of remission were 7.5 and 3.8 months for the diet and corticosteroid groups, respectively.
Although it is recognized that such dietary manipulation is not easy to achieve and maintain for both the patient and nutritionist, there is evidence to advocate an exclusion and re-introduction diet following successful induction of remission with enteral nutrition, rather than an immediate return to a ‘normal’ diet. An alternative to an exclusion diet is the LOFFLEX diet (low-fat, fibre-limited exclusion diet).107 This involves weaning the patient on to a limited range of foods, well tolerated by most patients with Crohn's disease, for 2–4 weeks following the completion of enteral nutrition. This is then followed by a re-introduction regime, and has been reported to maintain a remission rate of 59% at 2 years.107
Another study has found that both elemental and polymeric diets reduce the steroid dependence in steroid-dependent patients with Crohn's disease in remission.108 This represents another treatment approach, but further studies are needed. A more experimental approach has been the use of a transforming growth factor-β2-enriched polymeric diet in children with Crohn's disease.109 This was associated with mucosal healing and the down-regulation of intestinal mRNAs of pro-inflammatory mediators. A transforming growth factor-β2-enriched formulation is now available commercially (Modulen IBD, Nestle, Croydon, UK).
Enteral nutrition in colitis
Currently, there are insufficient data to assess the efficacy of enteral nutrition in patients whose Crohn's disease is confined to the colon. Some studies have not identified the site of disease as a predictor of the successful induction of remission with enteral nutrition.96, 110, 111 This is an area that merits further studies. There is no evidence to support the use of enteral nutrition in ulcerative colitis.
n-3 Polyunsaturated fatty acid
The fish oil-derived n-3 polyunsaturated fatty acid inhibits leukotriene B4, a potent chemo-attractant and pro-inflammatory eicosanoid implicated in the pathogenesis of inflammatory bowel disease. The association of an aberrant fatty acid profile and attenuated antioxidant status with disease activity in Crohn's disease forms the rationale for a potential therapeutic role of n-3 polyunsaturated fatty acid.112, 113 Supplementation with a liquid formula containing antioxidant and n-3 fatty acid significantly improved serum antioxidant status, reduced the proportion of arachidonic acid and increased the proportion of eicosapentanoic acid and docosahexanoic acid in plasma phospholipid and adipose tissue, suggesting a shift to an anti-inflammatory phenotype.100 Supplementation with nine capsules of an enteric-coated fish oil preparation (equivalent to 2.7 g n-3 polyunsaturated fatty acid) has been shown to be efficacious in maintaining remission in Crohn's disease when compared with placebo.114 However, another randomized, placebo-controlled, multi-centre European study failed to demonstrate the efficacy of n-3 supplementation over placebo in the maintenance of remission in Crohn's disease.115 High-dose n-3 supplementation may be associated with an unpleasant taste, halitosis and other gastrointestinal side-effects.
There has been some debate regarding the benefit of n-3 polyunsaturated fatty acid in ulcerative colitis.116, 117 A recent randomized controlled trial failed to demonstrate any efficacy of essential fatty acid supplementation in the maintenance of remission in ulcerative colitis.118
Parenteral nutrition as primary therapy
Total parenteral nutrition as primary therapy for inflammatory bowel disease has been used on the basis that it affords bowel rest and corrects nutritional deficiencies.6, 119, 120 The hypothesis that ‘bowel rest’ following total parenteral nutrition in Crohn's disease may allow mucosal healing has been disputed by at least two prospective studies designed specifically to address this issue.121, 122 Furthermore, the complications associated with total parenteral nutrition do not justify its use for the purpose of induction of remission in Crohn's disease. However, total parenteral nutrition does have a role in the healing of post-operative enterocutaneous fistula that arises from surgical anastomosis or more complicated fistulas in Crohn's disease.120, 123, 124 There is no evidence to support the use of total parenteral nutrition as primary therapy in ulcerative colitis.
Although glutamine is considered to be a non-essential amino acid, there may be an increased requirement for glutamine in the catabolic state.125 Glutamine is also an important nutrient for enterocytes and lymphocytes. There has been some suggestion that a lack of glutamine in standard total parenteral nutrition preparations may adversely affect enterocyte metabolism, cause mucosal atrophy and promote bacterial translocation. In animal models, glutamine improves mucosal integrity, reduces bacterial translocation126 and preserves both intestinal and extra-intestinal immunoglobulin A levels.127 Although the supplementation of glutamine in total parenteral nutrition has resulted in improved nitrogen balance and amino acid profile in critically ill patients,125, 128 studies of glutamine in inflammatory bowel disease have been less encouraging. Dietary glutamine supplementation in patients with Crohn's disease is neither effective in restoring intestinal permeability,129 nor in inducing remission.130
Butyrate, pre- and pro-biotics
The past 10 years have witnessed a surge in evidence supporting an alternative approach to the therapeutics of inflammatory bowel disease via manipulation of the resident enteric microflora. This may be achieved with exogenous butyrate, pre- and pro-biotics. Increasing appreciation of the pivotal role of the enteric microflora in the maintenance of a healthy gut, and in the pathogenesis of inflammatory bowel disease,131–135 has served to fuel interest in this area.
Short-chain fatty acids, including butyrate, propionate and lactate, are generated in the colon as a consequence of bacterial fermentation of dietary fibre by luminal Bifidobacterium, Eubacterium and Lactobacillus species. Short-chain fatty acids are readily absorbed by the intestinal mucosa and are an important source of substrate for metabolism by colonocytes; they are trophic to the intestinal mucosa, stimulate water and sodium absorption in the colon, and induce enzymes that promote mucosal restitution.133, 136
A direct anti-inflammatory role for butyrate, the most extensively studied of the short-chain fatty acids, may be attributable to its inhibition of the transmigration of nuclear factor-κB into the nucleus and its subsequent binding to DNA, thus preventing the transcription of pro-inflammatory mediators.137 In addition to their effect on epithelial cells, short-chain fatty acids have also been shown to inhibit lymphocyte activation and proliferation133 and myeloperoxidase activity in neutrophils, which directly causes tissue destruction.138
Butyrate enemas have been shown to be of benefit in the management of ulcerative colitis (reviewed in Han et al.6 and Scheppach et al.139). Alternative strategies of delivering short-chain fatty acids to the inflamed colon are by providing a substrate, a ‘pre-biotic’, for short-chain fatty acid production by colonic bacteria, or directly delivering pro-biotics to the intestinal lumen. In animal and human studies, ingestion of ‘resistant’ fibre has resulted in an increase in the population of Bifidobacillus and Lactobacillus in the colon and an increase in faecal butyrate concentration.140–145
A germinated barley foodstuff which contains hemicellulose-rich fibre and glutamine-rich protein has been studied by Japanese investigators.140–146 It has been shown to attenuate inflammation in dextran sodium sulphate, trinitrobenzene sulfonic acid (TNBS) and HLA transgenic animal models of colitis.140–142 Improvement in clinical and endoscopic indices has been reported in a pilot study of 10 patients with active ulcerative colitis with 4 weeks of treatment with germinated barley foodstuff.143 A lack of significant adverse events and good tolerability were also cited. Furthermore, the fibre may also firm up stool consistency due to its high water-holding capacity and possibly its adsorption of luminal bile salts.144, 145 An open-label, parallel-group, randomized study of another fibre, Platago ovata seeds, also reported equal efficacy to mesalazine in maintaining remission in patients with ulcerative colitis.146 More studies in this exciting area are needed.
Attention to nutrition-related issues must be considered central to the management of patients with inflammatory bowel disease. Clinical care pathways of inflammatory bowel disease should foster close involvement of dietitians in the assessment, counselling, treatment and follow-up in both the in- and out-patient setting. The use of nutrition as primary treatment in inflammatory bowel disease may attract wider support and research with increasing awareness of the harm of the long-term use of corticosteroids. Enteral nutrition may also be viewed as a form of neutraceutical approach to immunotherapy that is rapidly evolving. A ‘Crohn's disease-specific formula',147 comprising enteral nutrition, pre- and pro-biotics and growth factors, may offer a solution that fulfils the delicate balance between efficacy and toxicity.
We are grateful to Ruth Shirley, Research Dietitian at the Department of Gastroenterology and Clinical Medicine, Trinity College Dublin, for her invaluable advice on the manuscript.