Evidence-based dietary management of functionalgastrointestinal symptoms: The FODMAP approachPeter R Gibson and Susan J ShepherdMonash University Department of Medicine, Box Hill Hospital, Box Hill, Victoria, AustraliaAbstractBackground and Aim: Functional gastrointestinal symptoms are common and their managementis often a difficult clinical problem. The link between food intake and symptominduction is recognized. This review aims to describe the evidence base for restrictingrapidly fermentable, short-chain carbohydrates (FODMAPs) in controlling suchsymptoms.Methods: The nature of FODMAPs, their mode of action in symptom induction, results ofclinical trials and the implementation of the diet are described.Results: FODMAPs are widespread in the diet and comprise a monosaccharide (fructose),a disaccharide (lactose), oligosaccharides (fructans and galactans), and polyols. Theiringestion TR90 increases delivery of readily fermentable substrate and water to the distal

smallintestine and proximal colon, which are likely to induce luminal distension and inductionof functional gut symptoms. The restriction of their intake globally (as opposed to individually)reduces functional gut symptoms, an effect that is durable and can be reversed bytheir reintroduction into the diet (as shown by a randomized placebo-controlled trial). Thediet has a high compliance rate. However it requires expert delivery by a dietitian trainedin the diet. Breath hydrogen tests are useful to identify individuals who can completelyabsorb a load of fructose and lactose so that dietary restriction can be less stringent.Conclusions: The low FODMAP diet TR90 provides an effective approach to the managementof patients with functional gut symptoms. The evidence base is now sufficiently strong torecommend its widespread application.Functional gastrointestinal disorders (FGID) are very common andpresent as major challenges for clinicians, particularly as pharmaceuticaltherapies offer little more than mild palliation in the vastmajority of patients. The symptoms can markedly interfere withquality of life and rank second in the causes of absence from workor school.1 While the predominant underlying cause of symptomsappears to reside in the enteric nervous system, manifesting asvisceral hypersensitivity and/or motility disturbances, multipleother factors contribute to symptoms generation, including psychologicalfactors and diet. Consequently, treatment has spannedmultiple modalities and has involved a variety of health professionals,including medical practitioners, psychologists,

hypnotherapists,dietitians and naturopaths, each bringing a differentperspective. A major limitation has been the limited evidence basefor many therapies, not helped by the considerable placeboresponse seen in these disorders. However, dietary therapy, specificallythe low FODMAP diet TR90 (see below for explanation), hasnow emerged as a key player with a well-substantiated mechanismof action and evidence-based efficacy. This review will describethe theoretical basis for the diet, the evidence for efficacy and itsimplementation, and it will address unanswered questions.Mechanistic basis fordietary interventionThe physiological basis for the genesis of many functional gutsymptoms is luminal distension. Evidence for this comes frombarostat and gas infusion studies.2,3 Luminal distension not onlyinduces the symptoms of pain, the sensation of bloating and visibleabdominal distension, but may also lead to secondary motilitychanges. Thus, minimizing the consumption of dietary factors thatcan distend the intestine would theoretically lead to improvementin global symptoms that characterize FGID. In the case of two ofthe most common types of FGID involving the intestine, irritablebowel syndrome (IBS) and functional bloating, the distal smalland proximal large intestine would be the target regions of the gut.The intestinal lumen can be distended by solids, liquids and gas.Solids can be altered in the proximal large intestine by changingthe dietary fiber content both directly and indirectly via expansionor contraction of the bacterial mass. The liquid content in the distalsmall intestine will be dictated by the osmotic load in the lumen,and in the proximal large intestine by the osmotic load and theabsorptive ability of the epithelium. The gas content willdoi:10.1111/j.1440-1746.2009.06149.x252 Journal of Gastroenterology and Hepatology 25 (2010) 252–258© 2010 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltdinclude a component of swallowed nitrogen, but the majority willbe locally produced by bacterial fermentation. The volume that thegas creates will depend upon the number of molecules and itsdiffusion capacity across the epithelium and into the circulation.

Dietary TR90 components that will putatively lead to luminal distensionin the regions of interest will therefore be poorly absorbed in theproximal small intestine, will be small molecules (i.e. osmoticallyactive), will be rapidly fermented by bacteria (with the potential tobe fermented by small intestinal as well as cecal bacteria and toexpand the bacterial population), and will be associated withhydrogen rather than methane production. Dietary FODMAP arethe best fit for these principles.FODMAPsThe acronym, ‘FODMAP’—Fermentable Oligo-, Di- and Monosaccharidesand Polyols—was coined to describe a previouslyunrelatedgroup of short-chain carbohydrates and sugar alcohols(polyols).4 They comprise fructose, lactose, fructo- and galactooligosaccharides(fructans, and galactans), and polyols (such assorbitol, mannitol, xylitol and maltitol) all of which putativelyhave three common functional properties:• Poorly absorbed in the small intestine: Poor absorption occursby virtue of slow, low-capacity transport mechanisms across theepithelium (fructose), reduced activity of brush border hydrolases(lactose), lack of hydrolases (fructans, galactans), or moleculesbeing too large for simple diffusion (polyols).• Small and therefore osmotically-active molecules: This effecthas been demonstrated with, for example, a synthetic FODMAP,lactulose, which exerts a laxative effect when given in sufficientdose by increasing the liquidity of luminal contents and subsequentlyaffecting gut motility.5• Rapidly fermented by bacteria: The rapidity of fermentation bybacteria is dictated by the chain length of the carbohydrate;oligosaccharides and sugars are very rapidly fermented comparedwith polysaccharides such as soluble dietary fibre.6These functional properties have recently been confirmed instudies in which diets high and low in FODMAPs (rather than pureindividual FODMAPs) have been fed to volunteers. In a study of 10ileostomates, changes in dry-weight ileostomy effluent could beexplained entirely on the basis of dietary FODMAPs and theeffluent volume increased by a mean of 22% on the high FODMAPdiet.7 Fermentation of FODMAPs in the small intestine was suggestedby the recovery of only 34% of FODMAPs consumed in theileostomy effluent, although some fermentation in the ileostomybag ex vivo also would have contributed.

When the diets were fed tohealthy volunteers, breath hydrogen production, a marker of gasproduction in the intestine, was markedly elevated throughout theday.8 Furthermore, in methane-producers, high FODMAP intakefavored production of hydrogen over methane, which occupies asmaller volume per hydrogen molecule generated. Thus, all theputative functional properties have been confirmed to occur in vivoin association with dietary intake of FODMAPs.There is considerable evidence that individual FODMAPs induceabdominal symptoms. Acute provocation tests with lactose,9fructose9–11 fructo-oligosaccharides (FOS)12,13 or sorbitol9,14–17cause abdominal symptoms such as bloating, pain, nausea anddisturbed bowel habit (diarrhea and/or constipation) in manypeople, especially those with IBS.15 The role of lactose andpolyols in the induction of gut symptoms has been well-described inclinical practice; the dietary regimen for the management of lactosemalabsorption has been comprehensively addressed18 and mandatorydeclaration of ‘excess consumption may have a laxative effect’is in place for food products containing polyols. Increased flatulenceand change of bowel habits after consuming ‘windy vegetables’,such as lentils and baked beans, are common knowledgealthough identification of galactans, in addition to resistant starch,as the culprit molecules may not be. Additive effects fructose andsorbitol10,19,20 and lactose and fructans21 on abdominal symptomsare also well-described.The FODMAP TR90 concept in themanagement of functionalgut symptomsThere are two key components to the FODMAP concept.• The dietary approach restricts FODMAP intake globally, notindividually. Restriction of individual FODMAPs has been usedwith varying success in the management of functional gut symptomsfor a long time.

The best example is restriction of dietarylactose in patients with hypolactasia. Restriction of fructose,with or without sorbitol, has also been reported. However, suchapproaches have not become widespread in their application,perhaps in part related to their limited success. Restricting oneFODMAP in isolation ignores the likelihood that there is potentiallya range of FODMAPs in the diet, all of which have similarend-effects in the bowel. The innovation in the FODMAPconcept is that global restriction should have a far greater andmore consistent effect than limited restriction. Thus, the centralfocus is to reduce the intake of all poorly absorbed short chaincarbohydrates to be more effective in reducing luminal distensionthan merely concentrating on one of these. Such a globalapproach to restricting carbohydrates that have similar actions(high osmotic effect and rapid fermentation) should optimizesymptom control in patients with IBS.• FODMAPs do not cause the underlying FGID, but represent anopportunity for reducing symptoms. This concept is important asit steers away from the more traditional concepts of lactose‘intolerance’ versus ‘malabsorption’ and fructose ‘intolerance’versus ‘malabsorption’.

The reason the symptoms are triggeredby the ingestion of lactose or fructose in the individual is theresponse of the enteric nervous system to luminal distension(due to visceral hypersensitivity, excessive gas production due tothe nature of the resident microbiota, or motility problems withclearance of the fluid/gas) not because the malabsorption of thesugar is abnormal or a ‘condition’. After all, delivery of dietaryFODMAP to the distal small and proximal large intestine is anormal phenomenon, one that will generate symptoms if theunderlying bowel response is exaggerated or abnormal.FODMAPs in the dietWhile all FODMAPs are potentially important in the genesis ofsymptoms (summary of food sources of FODMAP are listed inTable 1), the relative contribution of different subgroups ofFODMAPs varies across ethnic and dietary groups due to the dosedelivered in the diet. In North American and Western Europeandiets, fructose and fructans are by far the most widespread