The scope of the problem

Adverse reactions to food are acknowledged by 5% to 45% of the general population, and GI complaints are predominant in approximately one-third to one-half of those affected. Offending foods are often referred to as trigger foods, dietary triggers, or culprit foods, sometimes leading to a nutritionally inadequate diet. Although food intolerance is a common perception among the general population, it can be demonstrated in only a relatively small proportion of the population when double-blind food elimination and challenge studies are employed. In a population study of food intolerance of 20,000 patients, 20.4% complained of food intolerance. Of the 93 subjects who entered the double-blind, placebo-controlled food challenge, 11.4%-27.4% had a positive reaction, with estimated prevalence of reactions to the 8 foods tested varying from 1.4% to 1.8% depending on the method of testing used. Women perceived food intolerance more frequently and showed a higher rate of positive results to food challenge.

Among individuals with IBS, 20% to 67% complain of subjective food intolerance, which is more prevalent than similar reports in matched controls. One population-based study reported a prevalence rate of perceived food intolerance of greater than 50% among subjects with IBS, a rate that was 2-fold greater than that reported by those without IBS. Another found that patients in a GI clinic with a final diagnosis of a functional disorder were four times more likely to report food allergies (FAs) or adverse reaction to food. The likelihood of a patient’s symptoms being functional increased even further if adverse reactions to both drugs and foods were reported. Such complaints seem to correlate with female gender and anxiety level in those with functional GI disorders.

Many patients suffering from IBS report an association of symptoms with specific foods. Although the foods that induce symptoms may be specific, the associated symptoms are often nonspecific and consistent with functional disorders, such as IBS. Many patients identify specific trigger foods (most commonly dairy, fructose, wheat products, and caffeine), but there is little evidence that IBS patients with food-related complaints are suffering from a true FA. Although there are undoubtedly some patients with IBS symptoms who suffer from an FA, the proportion of the total population of IBS sufferers with a true FA is small. Alternatively, there is likely to be substantial overlap between food intolerances/sensitivities and IBS, because these syndromes often have similar clinical presentations.

It is not uncommon for IBS patients to experiment with their diet or limit their diet before seeking medical attention. The foodstuffs most commonly implicated are wheat, corn, dairy products, coffee, tea, and citrus fruits. In a 2005 population-based sample comparing dietary consumption of specific food items and nutrients between individuals with IBS or dyspeptic symptoms and those without symptoms, no differences were seen in the consumption of frequently suspected culprit foods. In a survey of more than 1200 individuals with IBS, 63% were interested in knowing which foods to avoid. The lifestyle changes they had made or considered for treatment of IBS included small meals (69%); avoiding fatty foods (64%); higher fiber intake (58%); and avoiding milk products (54%), carbohydrates (43%), caffeine (41%), alcohol (27%), and high-protein foods, such as meats (21%).

How does food cause IBS symptoms?

Food processing is the primary function of the gastrointestinal (GI) tract, and food ingestion, through mechanosensation and nutrient sensing, causes major changes in GI sensorimotor and secretory function, which may contribute to symptom generation in IBS. On ingestion of nutrients, the proximal stomach relaxes and upper GI motility switches from interdigestive to fed state motility, characterized by phasic contractions in the antrum and the small bowel. In addition, food ingestion stimulates gastric and pancreatic secretion and is associated with the release of several anorexigenic hormones, such as cholecystokinin (CCK) and glucagon-like peptide-1 (GLP-1). Hence, food can influence symptom generation from the GI tract by stimulation of mechano- or chemoreceptors, activation of motor reflexes pathways, by altered secretion, and by (colonic) fermentation.

Most investigators agree that the stomach does not meaningfully detect the nutrient or caloric composition of a meal and that mechanical distention by the meal is the major gastric sensory signal upon food intake. Tension-sensitive mechanoreceptors in the stomach are involved in detecting the ingested nutrient volume and in triggering relaxation of the proximal stomach, which enables storage of the meal without a rise in intragastric pressure. Probably through similar mechanisms, the volume of the meal in the stomach is a trigger for the initial phase of the gastrocolonic reflex, a meal-induced increase in tone, and phasic contractility of the colon. The early (gastric) phase of colonic contractile activity in response food intake is mediated by activation of gastric mechanoreceptors and can be reproduced by distention of a gastric balloon, and suppressed by atropine. A later (intestinal) phase depends on duodenal delivery of nutrients.

Passage of nutrients to the small intestine activates a different set of neural and hormonal responses. The small intestine is well equipped to sense the presence of nutrients and their composition. Specialized receptors on enteroendocrine cells detect the presence of specific food constituents, pH, and osmolarity of small intestinal contents, and through altered release of peptides and other signaling molecules, neural and hormonal pathways are activated that may influence GI sensorimotor function and afferent signaling. In general, the presence of nutrients, gastric acid, or hyperosmolar fluids in the duodenum activates neurohormonal pathways that slow further delivery of gastric contents to the small intestine through inhibition of gastric tonic and phasic contractions and through closure of the pylorus. For instance, lipids in the duodenum activate long-chain free fatty acid receptors GPR120 and possibly GPR40 and release CCK from I-cells. CCK is a potent enhancer of colonic motility and of GI motility and has been implicated in the intestinal phase of the gastrocolonic reflex, based on observations in a dog model. The gastrocolonic reflex is exaggerated in IBS patients as a group, and this reflex has been implicated in the early postprandial increase in IBS symptoms that is present in the majority of patients. In humans, however, the colonic response to food ingestion was not inhibited by the CCK-A receptor antagonist loxiglumide. Recent studies have shown an exaggerated increase in rectal sensitivity to distention after ingestion of a meal in IBS. Because lipids were the strongest stimulus in these studies, it has been suggested that CCK might be involved, and exogenously administered CCK was shown to enhance rectal sensitivity in health.

Intestinal nutrient sensing is extensive, and several nutrient-related components have the ability to activate afferent pathways, alter GI reflex activity, or change the release of GI peptides. The presence of glucose in the lumen induces release of GLP-1 and serotonin, among other gut hormones, from enteroendocrine cells, through the sodium-glucose cotransporters, SGLT1 and SGLT3. These peptides may alter visceral sensorimotor function and could contribute to food-induced symptom aggravation. More recently, it has been shown that the small intestine also expresses the T1R and T2R G-protein–coupled taste receptors families. Taste receptors interact with specific Gα subunits, including α-gustducin, a taste-specific signaling protein. Because immunohistochemical studies have shown expression of α-gustducin in several types of enteroendocrine cells, it has been hypothesized that activation of taste receptor signaling molecules may alter the release of GI peptides, such as PYY, GLP1, or CCK. In turn, these hormones may alter endocrine secretions and appetite, but they may also affect motor behavior and sensory signaling, thereby contributing to changes in symptoms induced by ingestion of a meal. Finally, the gut also expresses members of the superfamily of transient receptor potential cation channels on enteroendocrine cells and afferent nerves. These may sense aspects of luminal contents, such as acidity, or the presence of food components, such as menthol or capsaicin, which may alter GI sensorimotor function through activation of neural or humoral pathways. The presence of taste receptors and transient receptor potential channels and their coupling to release of functionally relevant GI signaling molecules or activation or neural pathways could potentially help explain some of the intolerances to specific foods or tastes that IBS patients report and which cannot be explained by food allergies.

The physical properties of nutrients may also contribute to meal-related symptom aggravation, without involvement of sensing of nutrient composition. Nonabsorbable components of ingested food may also contribute to symptom generation and perception in IBS, either through changes in motility and transit that they induce or through fermentation by bacterial flora in the colon. Osmotically active substances in the small intestine, such as nonabsorbable sugars, attract fluids and this may lead to increased intestinal contractility, accelerated colonic transit, increased flatulence, and diarrhea. It is conceivable that such responses may be aggravated in IBS patients, where small intestinal transit and water content may already differ from health. Fiber, especially insoluble, may promote bacterial fermentation in the large intestine and symptoms of bloating and distention. Because the colonic flora may differ in IBS from health, these processes could potentially be altered in IBS, but additional studies are needed.

In summary, food can influence symptom generation from the GI tract through many pathways. These include activation of mechanoreceptors by the volume and physical properties of the meal and activation of chemoreceptor-activated pathways through a multitude of nutrient-sensing mechanisms. Nonabsorbable components may influence GI function and sensations through osmotic actions and colonic fermentation. There is emerging evidence of some altered nutrient-related GI functions in IBS, which include an exaggerated gastrocolonic reflex, enhanced sensitization to rectal distention after lipid ingestion, and changes in colonic bacterial flora that may be relevant for fermentation of nonabsorbable remnants. Detailed studies investigating potentially altered nutrient sensing in IBS are still lacking.

How does food cause IBS symptoms?

Food processing is the primary function of the gastrointestinal (GI) tract, and food ingestion, through mechanosensation and nutrient sensing, causes major changes in GI sensorimotor and secretory function, which may contribute to symptom generation in IBS. On ingestion of nutrients, the proximal stomach relaxes and upper GI motility switches from interdigestive to fed state motility, characterized by phasic contractions in the antrum and the small bowel. In addition, food ingestion stimulates gastric and pancreatic secretion and is associated with the release of several anorexigenic hormones, such as cholecystokinin (CCK) and glucagon-like peptide-1 (GLP-1). Hence, food can influence symptom generation from the GI tract by stimulation of mechano- or chemoreceptors, activation of motor reflexes pathways, by altered secretion, and by (colonic) fermentation.

Most investigators agree that the stomach does not meaningfully detect the nutrient or caloric composition of a meal and that mechanical distention by the meal is the major gastric sensory signal upon food intake. Tension-sensitive mechanoreceptors in the stomach are involved in detecting the ingested nutrient volume and in triggering relaxation of the proximal stomach, which enables storage of the meal without a rise in intragastric pressure. Probably through similar mechanisms, the volume of the meal in the stomach is a trigger for the initial phase of the gastrocolonic reflex, a meal-induced increase in tone, and phasic contractility of the colon. The early (gastric) phase of colonic contractile activity in response food intake is mediated by activation of gastric mechanoreceptors and can be reproduced by distention of a gastric balloon, and suppressed by atropine. A later (intestinal) phase depends on duodenal delivery of nutrients.

Passage of nutrients to the small intestine activates a different set of neural and hormonal responses. The small intestine is well equipped to sense the presence of nutrients and their composition. Specialized receptors on enteroendocrine cells detect the presence of specific food constituents, pH, and osmolarity of small intestinal contents, and through altered release of peptides and other signaling molecules, neural and hormonal pathways are activated that may influence GI sensorimotor function and afferent signaling. In general, the presence of nutrients, gastric acid, or hyperosmolar fluids in the duodenum activates neurohormonal pathways that slow further delivery of gastric contents to the small intestine through inhibition of gastric tonic and phasic contractions and through closure of the pylorus. For instance, lipids in the duodenum activate long-chain free fatty acid receptors GPR120 and possibly GPR40 and release CCK from I-cells. CCK is a potent enhancer of colonic motility and of GI motility and has been implicated in the intestinal phase of the gastrocolonic reflex, based on observations in a dog model. The gastrocolonic reflex is exaggerated in IBS patients as a group, and this reflex has been implicated in the early postprandial increase in IBS symptoms that is present in the majority of patients. In humans, however, the colonic response to food ingestion was not inhibited by the CCK-A receptor antagonist loxiglumide. Recent studies have shown an exaggerated increase in rectal sensitivity to distention after ingestion of a meal in IBS. Because lipids were the strongest stimulus in these studies, it has been suggested that CCK might be involved, and exogenously administered CCK was shown to enhance rectal sensitivity in health.

Intestinal nutrient sensing is extensive, and several nutrient-related components have the ability to activate afferent pathways, alter GI reflex activity, or change the release of GI peptides. The presence of glucose in the lumen induces release of GLP-1 and serotonin, among other gut hormones, from enteroendocrine cells, through the sodium-glucose cotransporters, SGLT1 and SGLT3. These peptides may alter visceral sensorimotor function and could contribute to food-induced symptom aggravation. More recently, it has been shown that the small intestine also expresses the T1R and T2R G-protein–coupled taste receptors families. Taste receptors interact with specific Gα subunits, including α-gustducin, a taste-specific signaling protein. Because immunohistochemical studies have shown expression of α-gustducin in several types of enteroendocrine cells, it has been hypothesized that activation of taste receptor signaling molecules may alter the release of GI peptides, such as PYY, GLP1, or CCK. In turn, these hormones may alter endocrine secretions and appetite, but they may also affect motor behavior and sensory signaling, thereby contributing to changes in symptoms induced by ingestion of a meal. Finally, the gut also expresses members of the superfamily of transient receptor potential cation channels on enteroendocrine cells and afferent nerves. These may sense aspects of luminal contents, such as acidity, or the presence of food components, such as menthol or capsaicin, which may alter GI sensorimotor function through activation of neural or humoral pathways. The presence of taste receptors and transient receptor potential channels and their coupling to release of functionally relevant GI signaling molecules or activation or neural pathways could potentially help explain some of the intolerances to specific foods or tastes that IBS patients report and which cannot be explained by food allergies.

The physical properties of nutrients may also contribute to meal-related symptom aggravation, without involvement of sensing of nutrient composition. Nonabsorbable components of ingested food may also contribute to symptom generation and perception in IBS, either through changes in motility and transit that they induce or through fermentation by bacterial flora in the colon. Osmotically active substances in the small intestine, such as nonabsorbable sugars, attract fluids and this may lead to increased intestinal contractility, accelerated colonic transit, increased flatulence, and diarrhea. It is conceivable that such responses may be aggravated in IBS patients, where small intestinal transit and water content may already differ from health. Fiber, especially insoluble, may promote bacterial fermentation in the large intestine and symptoms of bloating and distention. Because the colonic flora may differ in IBS from health, these processes could potentially be altered in IBS, but additional studies are needed.

In summary, food can influence symptom generation from the GI tract through many pathways. These include activation of mechanoreceptors by the volume and physical properties of the meal and activation of chemoreceptor-activated pathways through a multitude of nutrient-sensing mechanisms. Nonabsorbable components may influence GI function and sensations through osmotic actions and colonic fermentation. There is emerging evidence of some altered nutrient-related GI functions in IBS, which include an exaggerated gastrocolonic reflex, enhanced sensitization to rectal distention after lipid ingestion, and changes in colonic bacterial flora that may be relevant for fermentation of nonabsorbable remnants. Detailed studies investigating potentially altered nutrient sensing in IBS are still lacking.

Food allergy versus food intolerance

Up to 25% of the general US population believes they have an FA. True FA, however, occurs in approximately 4% to 8% of children and 1% to 4% of adults in the United States. Although controversial, there are data to suggest that the prevalence of FA is increasing. Few children retain their FA into adulthood, which accounts for the decrease in prevalence between children and adults. Many studies demonstrate that 50% to 90% of presumed food allergies are not actually allergies but rather a food intolerance (some adverse reaction to food) or food aversion (refusal to eat a food because it is potentially hazardous, socially inappropriate, or distasteful).

True FA is an immunologically mediated, adverse reaction that occurs reproducibly after a susceptible individual ingests a specific food. Although nonfood allergies can be diagnosed by well-defined criteria, the diagnosis of FA is far more difficult. This is particularly true for the clinical spectrum of food-related allergic manifestations that occur in the digestive tract, because many symptoms, such as oral/throat itching, nausea, vomiting, abdominal pain, diarrhea, and constipation, are vague and can be mistaken for other conditions. The best-established food reactions are IgE-mediated (considered type I hypersensitivity) and non–IgE-mediated reactions (including type III hypersensitivity [IgG or IgM immune complex reactions] and type IV hypersensitivity [delayed-type or cell-mediated reactions]). (See Table 1 for a classification of pathophysiologic reactions to foods.)

The best-established mechanism in FA is due to the development of IgE antibodies against the offending food (most commonly cow milk, eggs, peanuts, tree nuts, seafood, and shellfish). The diagnosis of an IgE-mediated FA is made by a carefully taken case history, supported by the demonstration of IgE sensitization either by skin prick tests or in vitro tests, and confirmed by positive food challenge. If the reaction is not IgE mediated (as in some children with cow milk allergy), however, these tests are often unrevealing. A helper T cells type 2 cytokine pattern is prevalent in allergic individuals at the site of the gut-associated lymphoid and mucosal system. Because individuals can develop immunologic sensitization (as evidenced by the presence of allergen-specific IgE) to food allergens without having clinical symptoms on exposure to those foods, an IgE-mediated FA requires both the presence of sensitization and the development of specific signs and symptoms on exposure to that food. Sensitization alone is not sufficient to define FA. The formation of IgE antibodies on antigen exposure may not necessarily induce clinical manifestations and vice versa (the quiescence of atopic disease does not necessarily correlate with a decrease in antibodies). Clinicians, however, may use the detection of these antibodies to identify potential offending antigens, the clinical significance of which can only be confirmed by reproduction of symptoms after oral food challenge. Furthermore, the absence of such antibodies helps exclude the presence of IgE-mediated FA. A double-blind placebo-controlled food challenge is the gold standard for establishing the presence of FA. Reproduction of GI symptoms with a single-blind or open food challenge is suggestive but subject to a higher likelihood of placebo response.

The spectrum of food allergies also includes delayed-onset or cell-mediated diseases, such as eosinophilic GI diseases (esophagitis, gastroenteritis, and colitis), food protein–induced enterocolitis syndromes (occurring in young children), and food-induced atopic dermatitis. These may be T-cell or IgE mediated. Two specific presentations that also fall within the FA category are the pollen food syndrome and the latex-FA syndrome, which both present with cutaneous and GI complaints and may include bronchospasm and anaphylaxis.

Food intolerances are non–immune-mediated adverse reactions to food and may be due to factors within foods, such as pharmacologic agents (histamine, sulfites, and caffeine), enzyme deficiency of the host (lactase deficiency), host-specific metabolic disorders (galactosemia and alcohol intolerance), or idiosyncratic responses induced by an unknown mechanism. To date, there has been no widely accepted definition put forth by an internationally sanctioned organization. Symptoms may cover a wide spectrum, but are usually minor complaints, such as headaches. Pharmacologic reactions to food and food additives may be caused by vasoactive amines (dopamine, histamine, serotonin, phenylethylamine, and tryamine) in selected foods, and food additives, such as sulfites, tartrazine, and monosodium glutamate. Symptoms of this type of reaction usually manifest outside of the GI tract in the form of headache, asthma, and urticaria.

Food allergy and food intolerance in IBS

Many studies have tried to examine the prevalence of FA in patients with GI symptoms. In one study of patients with IBS and inflammatory bowel disease (IBD), 32% complained of adverse reactions to food as a cause of their GI symptoms. FA was suspected, however, according to several criteria in only 14% and could be confirmed by endoscopic allergen provocation and/or elimination diet and rechallenge in only 3.2% of patients. Similar results were reported by Dainese and colleagues who found no difference in the frequency of positive skin prick tests in IBS patients when comparing those with and without self-reported adverse reactions to food. Similarly, there was little consistency between the specific foods reported to cause intolerance and those resulting from the tests (11 of 80 patients, 13.7%).

The prevalence of IBS-like symptoms is increased in patients with allergic symptoms, such as rhinitis and asthma, when compared with nonallergic patients. Another study showed that patients with asthma have an increased prevalence of IBS relative to patients with other pulmonary disorders and healthy subjects. Minor histologic abnormalities in the GI tract of patients with atopic diseases have been documented. Likewise, the presence of atopic conditions is increased in patients with diarrhea-predominant IBS. This higher prevalence of allergic symptoms in unselected gastroenterology patients and the higher prevalence of IBS-like symptoms in allergic patients both suggest that atopic disease and GI complaints may be associated, independently of whether or not patients suspect food hypersensitivity as the main cause of their symptoms.

Some authors have posited that patients with atopic diseases have impaired gut barrier function as indicated by increased intestinal permeability. The potential association between atopic disease, mild intestinal mucosal inflammation, and IBS symptoms in patients with self-reported food hypersensitivity was recently explored by Lillestol and colleagues. Symptoms, skin prick tests, serum markers of allergy, and intestinal permeability were recorded in 71 adult patients with self-reported FH. Patients with self-reported food hypersensitivity had a high prevalence of IBS and atopic disease (93% and 61%, respectively). Atopic patients had increased intestinal permeability and density of IgE-bearing cells compared with nonatopic patients, but GI symptoms did not differ between groups. Some investigators have suggested atopic IBS as a new subgroup of IBS, with the intestinal mucosal mast cell as a possible pivotal pathophysiologic factor. For further discussion on the link between immune response and functional bowel disease, readers are directed to a recent comprehensive review.

Elevated levels of both food-specific serum IgE and IgG4 antibodies have been associated with food hypersensitivity–induced atopic conditions also, although the significance of elevated IgG4 antibody has been questioned. There is emerging evidence that these antibodies are pathophysiologically relevant in IBS. Zar and colleagues tested 108 IBS patients with IgG4 titers, skin prick tests, and IgE titers and compared with controls. Of 16 common food articles, there were significantly higher titers in the IBS group to wheat, soy, beef, pork, and lamb compared with controls but no difference in IgE titers or skin prick tests. No correlation was seen between the pattern of elevated IgG4 antibody titers and patient symptoms (ie, no difference among IBS subtypes).

The identification of patients with FH and IBS symptoms can be challenging, and a newer application of the basophil activation assay has been used in attempt to further characterize these patients. A recent report by Carroccio and colleagues evaluated the prevalence of food hypersensitivity to milk or wheat and the performance of this in vitro basophil activation assay for FH in patients with IBS symptoms. This test has previously been applied to allergy diagnoses, but the investigators used this technique to identify FH in a group of IBS patients. One hundred and twenty patients with IBS based on Rome II criteria completed questionnaires addressing their symptoms and any possible self-perceived FH. Patients underwent a 4-week diet that strictly eliminated milk, wheat, egg, tomato, and chocolate. Responders subsequently underwent a series of double-blind, placebo-controlled, 2-week oral food challenges with milk and wheat. Ultimately, 24 (20%) of the IBS patients had a positive food challenge to milk and/or wheat (16% both, 3% milk, and 2% wheat) and were diagnosed with FH. Patients with FH developed symptoms after a median of 3 days of food challenge and 50% had to discontinue the food challenge because of recurrent symptoms. The FH patients tended to carry a longer diagnosis of IBS and showed a higher frequency of self-perceived food intolerance. Still, the majority of patients with self-perceived food intolerance had a negative response to double-blind food challenge. A small subset of patients who denied a relationship between food and their symptoms had a positive response to double-blind food challenge. The in vitro basophil activation assay was more often abnormal than were the serum measurements of total IgE and serum food-specific IgE. Using double-blind food challenge as a gold standard, specificity of the in vitro basophil activation assay was 86% in the IBS subjects with no false-positive results identified in the healthy controls. There was a high frequency of false-positive results when patients with other chronic intestinal inflammatory diseases were tested. More work is needed to validate the accuracy of the basophil activation assay used in this study.

Useful information on the role of foods in GI immunoallergic reactions could come from the colonoscopic allergen provocation test developed by Bischoff and colleagues. Briefly, during colonoscopy, specific food allergens were found to trigger mucosal weal and flare reactions in 54 of 70 (77%) patients with abdominal symptoms suspected of being food related, whereas no reaction was observed in healthy volunteers. The clinical relevance of these findings, however, requires further study.

Elimination diets

Many investigators have investigated dietary manipulation as a potential treatment strategy for patients with IBS. In 1982, Jones and colleagues were among the first to demonstrate symptom improvement after an elimination diet in two-thirds of their IBS patients. Based on such results, they and others hypothesized that food intolerance was the major factor underlying the pathogenesis of IBS and that a therapy based on a diet that excluded certain foods could be effective in IBS patients Subsequent studies, however, failed to conclusively confirm this hypothesis, with some groups demonstrating food hypersensitivity in only IBS patients with associated atopic disease or IBS-D and still others finding little benefit from exclusion diets. Overall, response rates to exclusion diets have ranged from 15% to 71%, with dairy products, wheat, and eggs the most commonly implicated food items. Foods high in salicylates or amines may also be problematic for IBS patients.

The reasons that underlie the widely variable response rates with exclusion diets are unknown but may include the lack of standardized protocols among the studies, varying inclusion/exclusion criteria, and inconsistent definitions of food intolerance and positive food challenge. Furthermore, a placebo diet is difficult to incorporate in clinical trials and the high placebo response in such trials necessitates large sample sizes to achieve appropriate power. Well-designed studies are needed to evaluate the role of exclusion diets in patients with IBS symptoms.

Specific diets

Gluten Restriction

Celiac disease is an immune-mediated disease that occurs as a consequence of exposure to gluten, a storage protein in wheat, barley, and rye. In patients with celiac disease, dietary gluten induces an abnormal mucosal immune response in genetically susceptible individuals associated with the markers HLA-DQ2 and HLA-DQ8. Patients with classic celiac disease suffer a variety of GI symptoms, such as bloating, abdominal pain, and diarrhea, and have villous atrophy on small bowel mucosal biopsies. Patients with celiac disease typically experience rapid clinical and more gradual histologic improvements on a gluten-free diet. Recent literature from around the world suggests that approximately 4% of patients with IBS symptoms actually have celiac disease. Studies from the United States suggest that a smaller proportion, approximately 1%, of patients with typical IBS symptoms turn out to have biopsy-proved celiac disease. There are data to suggest that a larger proportion of patients with IBS symptoms, although not meeting criteria for the diagnosis of celiac disease, are gluten sensitive. Gluten sensitivity is a term used to refer to a heterogeneous group of conditions in which gluten leads to some adverse clinical, serologic, or histologic reaction, which improves with dietary gluten restriction. Such patients often experience GI symptoms that can be clinically indistinguishable from patients with celiac disease or IBS.

Gluten sensitivity encompasses a broad array of disorders ranging from latent celiac disease to FA/hypersensitivity to simple intolerance. There is some acceptance that persistent low-grade inflammation may be present in a subset of patients with IBS, and multiple driving factors have been proposed, including small bowel bacterial overgrowth, postinfectious causes, and immune-mediated responses to specific dietary constituents, such as gluten. Several proposed mechanisms for gluten sensitivity without celiac disease exist, including increasing intestinal permeability of tight junctions or stimulating lamina propria macrophages and leading to a proinflammatory cytokine milieu. Nonceliac disease subjects who are +HLA-DQ2 may possess innate cells in a state of heightened activation compared with subjects who are DQ2−.

It has been argued that some patients with positive celiac antibody tests and evidence of genetic predisposition (+HLA-DQ2 or -DQ8) but only subtle abnormalities or entirely normal small bowel histology eventually develop villous atrophy and, thus, belong to the spectrum of celiac disease. Wahnschaffe and colleagues described a group of patients reporting IBS-D symptoms with +HLA-DQ2, minimal immunopathologic changes on duodenal biopsies (increased intraepithelial lymphocytes [IELs]), elevated celiac disease–associated antibodies in duodenal aspirate, and increased IgA deposition in the intestinal villi. Such IBS-D patients experienced significant reductions in symptoms with a gluten-free diet. Subsequent studies have shown a high likelihood of response to GFD in patients with GI symptoms, abnormal celiac disease antibody testing and genetic markers DQ2/DQ8, but normal or minimal small intestinal mucosal lesions. In a follow-up study, Wahnschaffe and colleagues enrolled 41 patients with IBS-D, +HLA-DQ2/DQ8, and positive IgA and IgG antibodies to tissue tranglutaminase and/or gliadin. GI symptoms and antibody levels were followed after 6 months of a gluten-free diet. In IBS-D patients, celiac disease–associated serum IgG antibodies (37%) and HLA-DQ2 expression (39%) were significantly more frequent than in a control group of IBD patients (18% and 23%, respectively). After 6 months of a gluten-free diet, stool frequency and GI symptom score returned to normal values in 60% of IBS-D patients who were positive and in 12% who were negative for HLA-DQ2 and celiac disease–associated serum IgG ( P <.05). The investigators concluded that celiac disease–associated serum IgG and HLA-DQ2 expression may identify a subset of IBS-D patients who respond to a gluten-free diet.

It is also important to emphasize that the finding of increased IELs on small bowel biopsy is not specific for celiac disease. A variety of other conditions, including Helicobacter pylori infection, small intestinal bacterial overgrowth, IBS, diabetes mellitus, microscopic colitis, and the use of various medications, have been associated with this histologic finding. Thus, in the absence of abnormal celiac antibodies or HLA-DQ2/8, other causes for duodenal lymphocytosis should be sought. That being said, the proportion of IBS patients with the isolated finding of increased IELs on small bowel biopsy who respond to a gluten-free diet remains to be clearly defined. Similarly, the proportion of IBS patients with isolated abnormal celiac antibody test results who will improve on a gluten-free diet is also unknown. Table 2 may serve as a useful guide for identifying patients who may benefit from a trial of a gluten-free diet, depending on their symptoms, biopsies, genotype, and serologies.

Table 2

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related posts:

Inflammation and Microflora Centrally Acting Therapies for Irritable Bowel Syndrome Emerging Pharmacological Therapies for the Irritable Bowel Syndrome Inflammation and Microflora The Role of Diagnostic Testing in Irritable Bowel Syndrome Emerging Pharmacological Therapies for the Irritable Bowel Syndrome

Stay updated, free articles. Join our Telegram channel

Join