Fibre in Enteral nutrition
Ceri J Green

S A J Clin Nutr 2000 November Vol 13 No 4

Standard enteral formulas do not contain fibre. They are of low viscosity and can therefore be administered easily through fine-bore tubes. Originally there was thought to be some clinical advantages to their use.1 It is now becoming apparent, however, that addition of fibre to enteral formulas may have important effects in terms of improving bowel function, preventing or alleviating enteral feeding-related diarrhoea, and maintaining or improving gut structure and barrier function. For this reason several formulas, usually containing soy polysaccharide (SP) as the sole fibre source, have been introduced onto the market. However, different types of fibre vary in their qualitative and quantitative effects. It is therefore unlikely that a single fibre source will be able to provide the full range of benefits to the intestine. Formulas containing a mix of different fibre types, as consumed in normal healthy diets, may be more appropriate.

Fibre types and normal intakes
Definition and classification of fibre
There is still much debate regarding the most appropriate definition of fibre.2-4 The preferred definition from a nutritional perspective is based on non-digestibility of dietary components in the small intestine. These components reach the colon where they will be wholly or partly fermented by the colonic microflora and/or partly excreted in the faeces. This definition includes non-starch polysaccharides (NSP), inulin, fructo-oligosaccharides (FOS), resistant starch (RS) and lignin.2

NSPs contain up to several hundred thousand monosaccharide units. Different types of polysaccharides differ according to the number of monosaccharides linked together, the different types present, the order in the polymer chain, the types of linkages between the monosaccharides, the presence of branches from the backbone of the polymer, and the number of monosaccharides with acidic groups present (e.g. uronic acids).5 One form of categorisation is summarised in Table I.

Fructo-polysaccharides (e.g. inulin) consist primarily of linear chains of fructose with a degree of polymerisation (DP) of up to 60 or more. One end of the fructose polymer is occupied by a b-D-fructose and the other end by an - a-D- glucose. FOS differ from fructo-polysaccharides only in chain length. The strict definition of an oligosaccharide is a chain of units with a DP of 3 - 95 or 3 - 10.4 RS is defined as `the sum of starch and starch products of starch degradation not absorbed in the small intestine of healthy individuals'.6 Intrinsic factors influencing starch digestion include physical inaccessibility due to starch contained within undisrupted plant structures (RS type I) such as whole or partly milled grains or seeds, or starch in granules of partially crystalline form (RS type II) such as in raw potato and green banana, or retrograded starch (RS type III) such as in cooked cooled potato, bread and cornflakes. Extrinsic factors include the extent of chewing, transit time, concentration of amylase and other digestive enzymes present, pH, amount of starch and presence of other food components that may influence digestion.6 Lignin comprises a group of polyphenolic compounds of widely varying molecular weights. It contributes to the structural rigidity of the plant cell wall and is an inhibitor of microbial cell wall digestion.7 Two main methods have been developed for analysing dietary fibre: enzymatic gravimetric methods (e.g. the Association of Official Analytical Chemists (AOAC) procedure), and enzymatic chemical methods (e.g. the Englyst and Southgate procedures).6 The former method measures NSP, lignin and a portion of RS, as does the Southgate method, whereas the Englyst method does not measure RS or lignin. Other methods are required for the quantitative measurement of inulin, FOS, RS and lignin.8-10

Different fibres can be classified according to their solubility in a buffer solution at a defined pH, and/or their fermentability in an in vitro system. Arbitrary cut-off values can be selected to categorise fibres. For example, a fibre may be classified as fermentable if it is at least 60% fermented according to a specific method (e.g. Titgemeyer et al., 199111), and non-fermentable if it is less than 40% fermented by the same method. However, there are no universally accepted definitions of solubility and fermentability. Since most fibre types are at least partially fermented, it may be more appropriate to refer to them as well-fermented and less well (or more slowly) fermented. Examples of fibre types that are well-fermented are pectin, guar gum, acacia fibre (gum arabic), inulin and FOS. Less well-fermented types include cellulose, wheat bran, corn bran, oat hull fibre and some RS. In general, well-fermented fibres are soluble, while less well-fermented types are insoluble. However, there are some exceptions, e.g. SP (insoluble) is quite well-fermented.11

Fibre intake in self-selected diets
Most data in the literature regarding fibre intake in the normal diet relate to `total dietary fibre', estimated using food tables. Table II summarises intakes of NSP and the proportion of soluble and insoluble types,12-14 inulin and FOS,15-17 RS, 13,18,19 and lignin7,20 in Western diets. In vegetarians and in countries where starchy foods are the main staple, intakes of all types of fibre are likely to be much higher than this.20-23 Fibre recommendations for healthy populations suggest that fibre intakes should be increased by increasing consumption of cereals, fruits and vegetables,13,24-28 although the guidelines in some instances are rather vague with regard to types and amounts of fibre (Table III). The UK is the most specific, recommending an average intake of 18 g NSP/day.24 There are no published guidelines for intakes of inulin, FOS, RS or lignin.

Physiological effects of fibre
Fibre intake influences nutrient absorption, sterol metabolism, carbohydrate and fat metabolism, stool bulking and weight, and colonic fermentation. It also influences gut structure and gut barrier function, and may even have some impact on immune function. Not all types of fibre have the same qualitative or quantitative effects; the full range of health benefits of fibre can only be obtained by consumption of a variety of fibre sources.29 The physiological effects depend primarily on the physical properties of a fibre.30 Low fibre intake has been associated with many Western diseases, such as obesity, diabetes mellitus and gastro-intestinal disorders including constipation, diverticulitis and colon cancer. It is suggested that increasing the amount of fibre in the diet may play a role in reducing the risk of such diseases, and in some cases may have a therapeutic role. These aspects of fibre are discussed fully elsewhere.13,31 The main focus in this review is on bowel-related effects of fibre in relation to enteral feeding.

Colonic fermentation
The colon contains a large and diverse population (over 400 species at an estimated level of 1011 - 1012 bacteria/g) of almost exclusively anaerobic bacteria that produce polysaccharidases and other enzymes that are capable of digesting endogenous and dietary proteins and carbohydrates that escape digestion in the small intestine.32,33 Fermentation is characterised by a complex series of inter-related reactions that result in the formation of a variety of end-products including gases (methane, hydrogen, carbon dioxide) short chain fatty acids (SCFAs) (C2-C5 organic acids, as well as an increased bacterial mass. The extent of fermentation and range and nature of end-products depends on a number of factors (Table IV).11,33-43 The principal SCFAs are acetate, propionate and butyrate. They account for 83 - 95% of the total SCFA concentration in the large intestine, which ranges from approximately 60 to 150 mmol/l, with a molar ratio of acetate/propionate/butyrate of approximately 60:25:15.33,43 Between 220 and 720 mM SCFA are produced daily with a typical Western diet, representing metabolism of between 20 and 70 g substrate per day.33 Luminal concentrations are highest in the caecum and right colon where concentrations of microflora are also highest, and pH levels are lowest in the right colon (5.4 - 5.9), increasing distally to 6.6 - 6.9.33,43 Most butyrate (approximately 90%) and 10 - 50% propionate are metabolised in the colonic mucosa, while the remaining propionate and most acetate enter the portal vein. Propionate and to a certain extent acetate are metabolised in the liver, where they act as precursors of lipids and sugars or direct energy sources.5 Acetate is the major SCFA found in peripheral blood.

SCFAs have a number of properties that may be of importance in maintaining normal bowel structure and function and preventing or alleviating colonic-based diarrhoea (Table V).44-47 The type and amount of SCFAs can be manipulated by feeding different fibres and combinations of fibre.38,43 Certain types of fibre, such as SP, gums and oat bran yield more butyrate than other fibre types such as pectin, sugar beet fibre and pea fibre.11 More slowly fermented fibres, such as cellulose and wheat bran, also effectively enhance butyrate levels throughout the colon compared with more rapidly fermented fibres, such as guar gum.41,48,49 Resistant starch is a good source of butyrate compared with NSP,50 although different types of RS produce different profiles of SCFA.51 Fibre mixes rather than single fibre sources favour butyrate production36,38 and do not increase gas production.52 Mixes also influence the rate of breakdown in the colon, for example RS spares NSP breakdown to a certain extent,51 and lignin may have a similar effect.

Stool weight
Different types of fibre have different effects on stool weight owing to differing physico-chemical properties, especially particle size and chemical composition, which influence solubility and fermentability.53 There are four distinct effects of fibre in the colon (water holding, stimulation of bacterial proliferation, reduction in transit time and increased gas production), which act together to result in increased stool bulk, reduced transit time and increased stool weight and frequency (Fig. 1).23,53

Clinical benefits of fibre
Constipation and diarrhoea during enteral feeding
Constipation is not well defined owing to large inter-individual variations in bowel habit. In general, the term refers to low bowel frequency, long transit time, difficult stool expulsion and/or hard, dry stools. Diarrhoea can be defined as the passage of more than 200 g of stool per 24 hours on an average Western diet, based on physiological principles that take into consideration the capacity of the intestine to assimilate fluid and electrolytes.54 In routine practice it is difficult to measure stool weights, and therefore clinical definitions tend to be used, e.g. `abnormal looseness of stool'55 or `the passage of too-frequent stools or stools of too loose a consistency that are of inconvenience to the nursing staff and/or patient'.56

Constipation, faecal impaction and use of laxatives and other elimination aids are frequently cited as problems in the care of chronically sick, disabled and non-ambulatory populations.57-59 In the acute setting, the most obvious gastro-intestinal complaint associated with enteral feeding is diarrhoea.60,61 There are a number of adverse effects of diarrhoea apart from the obvious problem of loss of water, electrolytes and nutrients; these include discomfort of asking for/using bedpan, faecal incontinence, odour, effects on skin care, risk of contamination and increased costs (disposal, clean bed linen, renewing dressings, increased nursing time).62 The true incidence of enteral feeding-related diarrhoea is difficult to define because of the lack of a universally accepted definition.63

Effect of fibre-supplemented enteral formulas on constipation
Administration of a fibre-free enteral formula reduces bowel frequency and stool weight and increases transit time compared with a self-selected diet in healthy volunteers.64-69 A few studies have reported diarrhoea in healthy volunteers,70-73 but this may relate more to the method of feeding than the composition. Some of these studies also examined the effect of enteral formulas supplemented with single fibre sources. The most commonly tested fibre has been SP, with a few studies also examining the effects of cellulose, pectin, modified guar, soy oligosaccharide and oat fibre. Different types of fibre sources have different effects. Supplementation of liquid diet with cellulose alone resulted in hard stools and difficult elimination in healthy subjects,74 while modified guar, soy oligosaccharide and oat fibre administered separately had no effect on stool characteristics.66,68,75 Of the single fibre sources, SP seemed to show most benefit,65,67,68 although consistent and significant improvements were not demonstrated in all studies.

Patient studies with SP-enriched enteral formulas also only show trends in improvement. In a 1-year study of long- term fed patients,58 wet and dry stool weights and stool frequency were significantly increased, although the use of elimination aids was not reduced. Others59 found no differences in stool frequency or wet weight with addition of SP, although the wet stool weights on fibre-free diet were much higher than expected.56 Fischer et al.57 showed no effects on frequency or transit time, but demonstrated a trend in increasing stool weight to levels comparable with low normal values for healthy adults consuming a low fibre diet. Heymsfield et al.76 showed a tendency to increased stool weights with increasing SP intakes compared with patients who received fibre-free formula when fed for 1 - 2 week periods with each formula, although differences were not statistically significant. Few studies have yet been performed with mixed fibre sources, but early results are encouraging. In healthy volunteers, a formula supplemented with a mix of different fibre sources (Nutrison Multifibre, NV Nutricia, Zoetermeer, Netherlands) designed to reflect more closely the range and type of fibre habitually consumed in the normal diet (cellulose, SP, acacia fibre, inulin, FOS and RS) resulted in bowel frequency and transit time comparable with that during self-selected diet.69 Transit time was significantly improved compared with fibre-free formula, although stool weight was not. In stable medical patients, a combination of SP and oat fibre showed benefits in terms of stool frequency and weight.77

In conclusion, supplementation of fibre-free enteral diet with a single fibre source or mixed fibre appears to exert only marginal benefits on stool weight in healthy volunteers and long-term fed patients. It may be that it will be impossible to manufacture a formula that is of sufficiently low viscosity to pass through a fine bore nasogastric tube, while containing sufficiently large fibre particles to exert the expected effects on stool weight and sufficient fermentable fibre for adequate SCFA generation. Stool frequency and transit time may therefore be more appropriate parameters of bowel function in enterally fed patients.

Effect of fibre-supplemented enteral formulas on diarrhoea
Diarrhoea associated with enteral feeding is multifactorial and has been associated with lactose intolerance, formula temperature, osmolarity, protein sources, bacterial contamination, delivery method, drug therapy, starvation and hypoalbuminaemia. Many of these factors may be less of a problem than often thought, and most can be avoided or controlled. An often overlooked factor is a low fibre diet, which may result in diarrhoea because of inadequate generation of SCFAs. Animal studies have demonstrated that pectin and SP administration increase colonic water absorption, probably mediated via SCFA production.78 Antibiotics reduce SCFA production by interactions with the colonic microflora,79 and antibiotic-related diarrhoea is associated with low luminal SCFA.80 In addition, acute watery diarrhoea in patients with cholera is associated with a reduction in luminal SCFA and a cessation of net water absorption and a decrease in net sodium absorption in the colon.81 Rectal administration of SCFA at levels mimicking normal faecal concentrations (acetate/propionate/butyrate ratios of 60:40:20 mmol/l) reversed the defective absorption of water and sodium.

Further evidence for a role of SCFA in diarrhoea was the observation that the fluid secretion observed in the ascending colon during intragastric feeding of fibre-free enteral diet can be reversed by caecal infusion of physiological concentrations of acetate, propionate and butyrate (50:20:20 mmol/l).82 Liquid stools were reversed with addition of fibre (pectin) in healthy volunteers,70 and Duncan et al.73 suggested that enteral formula supplemented with a mixed fibre source may have a protective effect on colonic motor function. The earliest report of fibre as a means of reducing the incidence of enteral feeding-related diarrhoea was anecdotal,83 and attempts to substantiate this work have led to conflicting findings. Several studies have shown no benefit of fibre addition on incidence of diarrhoea,84-88 while others have shown a reduced incidence.89-91 Although the types and amounts of fibre used in these studies were similar (psyllium, SP and partially hydrolysed guar), there were large variations in study design, patient populations, definition of diarrhoea, and perhaps most importantly, the use of antibiotic therapy. None used combinations of fibre, which might be of more benefit than a sole fibre source, thus potentially allowing a greater selection of types of substrate for existing colonic flora to metabolise, thereby producing SCFAs along a greater length of the colon, improving water absorption and maintaining colonic integrity. A significant decrease in diarrhoea in neurological intensive care patients was seen with an SP and oat mix compared with a formula containing a low amount of SP, although no definition of diarrhoea was given.92

Effect of fibre-supplemented enteral formulas on gut-barrier function
Apart from its role in digestion, absorption and substrate redistribution, the gastro-intestinal tract constitutes a major immune organ and acts as a barrier to prevent entrance of microorganisms into the body. Changes in intestinal morphology, which occur in association with starvation and stress, are associated with changes in gut- barrier function leading to translocation of viable indigenous bacteria or bacterial products from the gastro- intestinal tract to the mesenteric lymph nodes and other organs.93 Bacterial translocation has been shown to occur in stressed animals and in clinical situations where the gut is damaged, e.g. surgery for Crohn's disease, intestinal obstruction and during bone marrow transplantation, although the mechanisms and clinical significance remain unclear.94 However, if the so-called `gut hypothesis' of translocation leading to systemic infection is proved, it will have enormous implications for concepts to improve gut integrity and enteral nutrition may become of primary importance in the treatment of gut dysfunction in critical illness.95 Although the provision of fibre-free enteral formulas is likely to have a beneficial effect on the preservation of upper small intestinal mass, ileal and colonic hypoplasia will occur in the absence of faecal bulk.96-98 Rodent experiments have demonstrated that fibre-free diets cause atrophy of the ileum and colon, and fibre-supplemented diets have a proliferative effect on the mucosa.99,100 Although further work is required in this area in order to define more closely the effects of different fibres or their combinations in the clinical setting, there is much preliminary evidence from animal studies to suggest that fibre, in particular a mixture of well-fermented and less well-fermented types, may have a beneficial effect on other components of the gut barrier, as well as the gut mucosa, and on bacterial translocation. The potential effects and possible mechanisms are summarised in Table VI.34,38,40,50,96,99-141 These effects may also be of great importance for protecting and repairing the gut, such as in patients with ulcerative colitis and ileal pouchitis, colonic anastomoses and short bowel syndrome, and critically ill patients.45,47,142-148

Tolerance to enteral formulas containing new fibre sources
Adults have wide inter-individual tolerance to different fibre types, and sudden introduction of a large amount of a particular type of fibre may cause unpleasant gastro-intestinal side-effects, for example flatulence, abdominal bloating, intestinal cramps, noise and pressure, and alterations in stool consistency. It is therefore frequently recommended that fibre intake be increased gradually, while ensuring adequate fluid intake. SP is the most common fibre source in enteral formulas and is well tolerated in healthy volunteers65,67,68 and patients in both the chronic57-59,76 and acute90,91,149 care settings. However, fibre formulations containing new ingredients such as inulin and FOS are now appearing on the market, and it is important to examine the tolerance of these ingredients since gastro-intestinal symptoms with high intakes (20 - 40 g) have been noted in healthy volunteers consuming normal diets.150,151

An enteral formula containing 30 g inulin/2 litres administered to 6 healthy volunteers resulted in diarrhoea in 3 and excess bloating and flatulence in all 6, but a mixture of inulin (15 g) and SP (15 g) was well-tolerated (D Silk - personal communication). Studies in stable patients using daily doses of over 30 g/day of inulin resulted in no obvious trend in stool consistency, but there was an obvious increase in flatulence.152 In healthy volunteers, an enteral formula containing a variety of fibres including inulin, FOS and RS, at levels reflective of those in the normal diet, was well-tolerated.69

Therefore, high intakes of well-fermented fibres such as inulin and FOS may well lead to some gastro-intestinal side-effects, which will vary widely in severity between individuals. However, in smaller quantities, more reflective of normal dietary intakes and taken in addition to other fibres, these substrates are of great interest as well-fermented fibre sources with good technological properties and possible specific effects on the intestinal microflora.

Effect of fibre on macronutrient and micronutrient absorption
Faecal nitrogen excretion is frequently increased during consumption of a high fibre diet.122,123,153 However, overall nitrogen retention is not compromised owing to a compensatory decrease in urinary nitrogen excretion which is explained primarily by microbial fermentation in the large intestine.122,123 A fibre-enriched diet may also increase faecal fat excretion by a variety of mechanisms.154,155 However, at modest levels of fibre intake, the small increase in energy loss during fibre supplementation is unlikely to be significant. Little effect of enteral nutrition supplemented with various single fibre sources or mixtures was seen on macronutrient absorption in healthy volunteers,64,156,157 ileostomists158 and patients.76 Furthermore, with prolonged use of fibre, adaptive mechanisms would be expected to reduce energy loss further.40,159 In addition, it may be that some of the energy lost is at least partially compensated for by the contribution of fibre fermentation to digestible energy.160 It is therefore very unlikely that the supplementation of enteral nutrition with fibre will lead to deleterious effects on macronutrient absorption in patients with normal gastro-intestinal function. However, further studies are required to examine the effects of fibre on nutrient absorption in patients following intestinal resection or with impaired pancreatic function.156

There has been some concern that fibre may impair mineral retention because of decreased bioavailability as a result of the cation exchange capacity of dietary fibre (mainly associated with unmethylated galacturonic acid residues in fruit and vegetable fibre and phytic acid in cereal brans) and/or the mineral complexing ability of phytate.161,162 SP does contain some phytate (0.8 - 1.2%) and enteral formulae containing SP as the sole source of fibre (about 15 g/l) will therefore contain approximately 0.02 g phytate/100 ml (0.4 g/2 000 ml). This is considerably lower than estimates of phytate intake in the normal diet (0.6 - 0.8 g/day).163 The amount of phytate reaching the small intestine is likely to be even lower than this, since phytate is partially broken down by heat processing and action of gastric acid.164 Adverse effects of phytate may have been overemphasised in the past,164,165 and it may even have some beneficial antioxidative properties.166 Fibres such as inulin, FOS, acacia fibre and RS contain even less phytate and uronic acid residues than SP.

Several animal studies demonstrate that mineral absorption is in fact improved during supplementation with a variety of different fibres, including inulin, RS, cellulose, SP and pectin.167-170 In healthy volunteers, slightly negative effects for some minerals and trace elements at higher doses of SP (e.g. 40 g) have been shown,167,171 but others have shown no negative effect76,172 or even improved absorption.172 Longer term studies may have shown better results. In conclusion, impaired mineral absorption is only likely to be of consequence with extremely high intakes of fibre and phytate, or when mineral and trace element intake is limited, which is not the case with fibre-enriched enteral formulas.

Conclusion
The theoretical benefits of fibre in enteral formulas for the purposes of maintaining or improving normal bowel structure and function are extensive. Until now, it has not proved possible to demonstrate these effects conclusively in clinical studies. This may be related to a number of reasons, including inadequate patient type and number, too brief study periods, inadequate use of objective parameters for assessing bowel function and diarrhoea, concomitant antibiotic therapy, and difficulties in assessing gut structure and barrier function in vivo. Possibly one of the most important reasons for the failure to demonstrate an improvement in bowel function or a reduction in diarrhoea is that in the majority of studies a sole fibre source was used. In view of the wide variety of different types of fibre that are habitually consumed as part of a normal diet, it can be hypothesised that a mixture of such fibres may be more effective, providing a greater selection of types of substrate to the colon. Combinations of fibres with differing fermentation characteristics might also be expected to have greater impact on gut morphology, barrier function and translocation of bacteria and endotoxin than single fibre sources. This suggests that in the future, fibre mixes may become a standard component of almost all enteral formulas, and ultimately it may even be possible to define optimal fibre mixes for specific patient groups, depending on the objectives of administration. However, since many effects of fibre are mediated via SCFAs, antibiotic therapy may render fibre administration only partially effective. In this case, reducing the use of unnecessary antibiotic therapy, rapidly re- establishing the intestinal microflora (e.g. using probiotics in addition to fibre)173 or finding a way of providing SCFA directly to the colon will be necessary. Inability to access the gut non-invasively and to assess translocation satisfactorily in patients suggests that other models will continue to be important for indicating the potential role of different fibre sources and combinations in influencing gut structure and barrier function.

This paper was presented as a keynote address at the 1997 SASPEN Congress.

References

1. Russell RI. Elemental diets: progress report. Gut 1975; 16: 68-79.
2. Bär, A (Bioresco AG). Definition of dietary fibre for nutrition labelling purposes: Conclusions of a food industry ad-hoc Working Group on Dietary Fibre. Binningen, 1993.
3. Asp N-G. Dietary fibre analysis - an overview. Eur J Clin Nutr 1995; 49: suppl 3, S42-S47.
4. Englyst HN, Quigley ME, Hudson GJ. Definition and measurement of dietary fibre. Eur J Clin Nutr 1995; 49: suppl 3, S48-S62.
5. Gurr MI, Asp N-G. Dietary Fibre. ILSI Europe Concise Monograph Series. Washington DC: ILSI Press, 1994.
6. Englyst HN, Cummings JH. Non-starch polysaccharides (dietary fibre) and resistant starch. Adv Exp Med Biol 1990; 270: 205-225.
7. Silk DBA. Fibre and enteral nutrition. Gut 1989; 30: 246-264.
8. Dysseler P, Hoffem D. Ring test 1993 - 1994 for total and resistant starch determination. In: Asp N-G, van Amelsfoort JMM, Hautvast JGAG, eds. Proceedings of the Concluding Plenary Meeting of EURESTA. Physiological Implication of the Consumption of Resistant Starch in Man. European FLAIR-Concerted Action No. 11 (Cost 911), Wageningen, The Netherlands: Euresta, 1994: 87-94.
9. Dysseler P, Hoffem D. Inulin, an alternative dietary fibre. Properties and quantitative analysis. Eur J Clin Nutr 1995; 49: suppl 3, S145-S152.
10. Theander O, Westurlund E. Determination of individual components of dietary fibre. In: Spiller GA, ed. Dietary Fiber in Human Nutrition. 2nd ed. Boca Raton: CRC Press, 1992: 77-98.
11. Titgemeyer EC, Bourquin LD, Fahey GC, Garleb KA. Fermentability of various fiber sources by human fecal bacteria in vitro. Am J Clin Nutr 1991; 5: 1418-1424.
12. Bingham SA, Pett S, Day KC. Non-starch polysaccharide intake of a representative sample of British adults. J Hum Nutr Diet 1990; 3: 333-337.
13. British Nutrition Foundation. Complex Carbohydrates in Foods. London: Chapman and Hall, 1990.
14. Emmett PM, Symes CL, Heaton KW. Dietary intake and sources of non-starch polysaccharide in English men and women. Eur J Clin Nutr 1993; 47: 20-30.
15. Teeuwen H, Thon‚ M, Vandorpe J. Inulin: a versatile fibre ingredient. International Food Ingredients 1992; 5: 10-14.
16. Roberfroid M, Gibson GR, Delzenne N. The biochemistry of oligofructose, a non-digestible fiber: an approach to calculate its caloric value. Nutr Rev 1993; 51: 137-146.
17. Van Loo J. Natural occurrence of inulin and oligofructose. In: Proceedings of the First ORAFTI Research Conference 2nd ed. Brussels: University of Louvain, 1995: 119-130.
18. Schweizer TF, Andersson H, Langkilde AM, Reimann S, Torsdottir I. Nutrients excreted in ileostomy effluents after consumption of mixed diets with beans or potatoes. II. Starch, dietary fibre and sugars. Eur J Clin Nutr 1990; 44: 567-575.
19. Dysseler P, Hoffem D. Estimation of resistant starch intake in Europe. In: Asp N-G, van Amelsfoort JMM, Hautvast JGAG, eds. Proceedings of the Concluding Plenary Meeting of EURESTA. Physiological Implication of the Consumption of Resistant Starch in Man. European FLAIR-Concerted Action No. 11 (Cost 911), Wageningen, The Netherlands: Euresta, 1994: 84-86.
20. Englyst HN, Trowell H, Southgate DAT, Cummings JH. Dietary fibre and resistant starch. Am J Clin Nutr 1987; 46: 873-874.
21. Khokhar S, Pushpanjali. Dietary fibre composition of Indian foods and intakes of Indian vegetarian populations. J Hum Nutr Diet 1995; 8: 447-456.
22. Nathan I, Hackett AF, Kirby S. The dietary intake of a group of vegetarian children aged 7 - 11 years compared with matched omnivores. Br J Nutr 1996; 75: 533-544.
23. Cummings JH. Anatomy and physiology of the human colon. In: Proceedings of the First ORAFTI Research Conference. Brussels: University of Louvain, 1995: 95-107.
24. UK Department of Health. Dietary Reference Values for Food Energy and Nutrients for the United Kingdom. Report on Health and Social Subjects, 41. London: HMSO, 1991.
25. Voedingsraad. Nederlandse Voedingsnormen 1989: advies opgesteld door de Commissie Voedingsnormen. 'sGravenhage: Voorlichtingsbureau voor de Voeding, 1989.
26. Nordic Working Group on Diet and Nutrition. Nordic Recommendations 1996. Scand J Nutr 1996; 4: 161-165.
27. National Research Council. Recommended Dietary Allowances. 10th ed. Washington, DC: National Academy Press, 1989.
28. Deutsche Gesellschaft für Ernährung. Kohlenhydrate, Ballastoffe. In: Empfehlungen für die Nährstoffzufuhr. 5. überarbeitung 1991. Frankfurt/Main und Umschau Verlag, 1991: 36-39.
29. Hazan A, Madar Z. Preparation of a dietary fiber mixture derived from different sources and its metabolic effects in rats. J Am Coll Nutr 1993; 12: 661-668.

Last updated: 17-Feb-2004