After individual presentations on various disorders of the gastrointestinal tract, issues of classification were discussed for consensus. As a beginning, the workshop participants agreed that the diagnosis of immunologic reactions to dietary products in the gastrointestinal tract in infants and young children is often made on the basis of one or more of the following criteria:
1. The history of an allergic (immunoglobulin [IgE]-mediated or non–IgE-mediated) or allergic-like hypersensitivity reactions to foods ingested or following exposure.
2. The exclusion of other anatomic, functional, or infectious reasons for such reactions.
3. The proven or presumed relationship to ingestion of a specific dietary product or food based on repeated hypersensitivity reactions, clinically, after either challenge or inadvertent re-exposure.
4. The finding of either positive immediate-reacting, IgE-mediated epicutaneous skin test reactions to commercial or fresh foods or in vitro identification of circulating IgE anti-food antibodies in proven or presumed IgE-mediated reactions.
5. The failure to respond to conventional therapies of gastrointestinal conditions of an anatomical, functional, or infectious nature.
6. The positive clinical response to avoidance of specific food or other dietary exposures, over time.
7. The positive clinical response to the use of medications, usually used to treat acute allergic or allergic-like conditions (e.g., epinephrine) by injection, to treat anaphylaxis or anaphylactoid reactions or allergic inflammation (e.g., corticosteroids).
8. The similarities of clinical manifestations of the reaction as well as other characteristics to clinical syndromes, either proven or presumed to be caused by immunologic mechanisms.
9. The lack of other explanations for this clinical allergic-like reaction.
In group discussions, it became clear that most attendees agreed that it is difficult to classify all the known or presumed types of immunologic gastrointestinal reactions to foods and other dietary products in infants and young children under a single unifying concept. From a mechanistic point of view, most reactions are presumed to be immunologic in nature, rather than proved to be so. The only clear-cut mechanism involves IgE anti-food antibodies and mast cell activation. The finding of eosinophils in a gastrointestinal biopsy is suggestive of a hypersensitivity reaction but does not in itself constitute a diagnosis of an immunologic reaction. Eosinophilic infiltrates detected in a tissue biopsy of the gastrointestinal tract may be due to systemic disease unrelated to a specific dietary-induced hypersensitivity-like clinical reaction or may play a passive role rather than an active role in such a clinical reaction. A response or absence of response to a medication or other therapies does not in itself constitute a diagnosis of these conditions, because these same therapies are often used to treat other gastrointestinal diseases and conditions unrelated to diet or immunologic reactions. The clinical syndromes presumed to involve immune-mediated gastrointestinal reactions are often based on clinical and/or laboratory and biopsy criteria, which overlap with other gastrointestinal disorders unrelated to diet or a proven immunologic mechanism.
Regardless of the immunologic mechanism involved, symptoms of gastrointestinal hypersensitivity are very similar, but generally vary in time of onset, severity, and persistence. The consensus conference reviewed the terminology applied to disorders of infants and children associated with immune responses to foods. After extensive discussions, the group agreed on the terminology and classification approach presented in Table 1. Each clinicopathologic entry is discussed in the following sections, along with the features that distinguish it from the others.
Immunoglobulin E-Mediated Disorders
Immediate gastrointestinal hypersensitivity is defined as an IgE-mediated gastrointestinal reaction that often accompanies allergic manifestations in other target organs (e.g., skin, lungs) and results in a variety of symptoms (1) (Table 2). A variety of evidence supports this definition. In early studies of IgE-mediated food hypersensitivity, radiographic changes were associated with food-allergic reactions. Rowe gave four food-allergic patients barium mixtures containing specific food allergens. He then studied them radiographically, observing gastric retention, hypermotility of the intestine, and colonic spasm (2). In fluoroscopic studies by Fries and Zeiger of 12 food-allergic children, the outcome of barium meals with and without food allergens were compared (3). The most prominent findings included gastric hypotonia and retention of the allergen test meal, prominent pylorospasm, and subsequent increased or decreased peristaltic activity of the intestines. “Immediate” allergic reactions in the stomachs of six food-allergic patients were also observed by gastroscopy (4). A small amount of the appropriate food allergen was placed on the gastric mucosa and then revisualized 30 minutes later. The mucosa then appeared markedly hyperemic and edematous with patches of thick gray mucus and scattered petechiae. Similar more recent studies have confirmed these earlier observations (5). In addition, biopsy samples obtained before and after challenges were found to have significant decreases in stainable mast cells and tissue histamine after positive food challenges consistent with IgE-dependent mast cell activation. Positive skin test or radioallergosorbent test (RAST) results to responsible food allergens are present in virtually all cases (<95%).
It was agreed that symptoms due to immediate gastrointestinal hypersensitivity typically develop within minutes to 2 hours of consumption of the responsible food allergen. These consist of nausea, abdominal pain, colic, vomiting within 1 to 2 hours, and/or diarrhea within 2 to 6 hours. In young infants, immediate vomiting is not always a consistent finding, and some of these infants vomit intermittently and show failure to thrive. In children with atopic dermatitis and food allergy, the repeated ingestion of food allergen appears to induce a partial desensitization of gastrointestinal mast cells resulting in subclinical reactions. Frequently, these children have poor appetite, poor weight gain, and intermittent abdominal pain, but carbohydrate permeability studies, a measure of gut wall integrity, demonstrate abnormal permeability in such patients (6,7).
Oral allergy syndrome (OAS) is estimated to affect up to 40% of adults who have pollen allergy, especially to birch, ragweed, and mugwort pollens (8) (Table 3). It is reported as a form of IgE-mediated contact allergy that is confined to the oropharynx, and rarely affects other target organs. This conference endorses this description. Local IgE-mediated mast cell activation provokes the rapid onset of pruritus, burning, tingling and swelling of the lips, tongue, palate and throat and, occasionally, a sensation of pruritus in the ears and/or tightness in the throat. Symptoms are recognized as usually short-lived and are most commonly associated with the ingestion of various fresh fruits and vegetables (9,10). Ragweed-allergic patients may experience OAS after contact with various fresh melons (e.g., watermelon, cantaloupe, honeydew) and bananas (10,11). Symptoms may be more prominent after the ragweed season, corresponding to the seasonal rise in ragweed-specific IgE levels. Birch pollen–allergic patients may have symptoms after the ingestion of raw potatoes, carrots, celery, apples, hazelnuts, and kiwi (10–13). Cross-reactivity between birch pollen and various fruits and vegetables is due to homology among various pathogenesis-related proteins important in the defense against plant diseases. Oral symptoms also have been described among several fruits in the Prunoideae subfamily, and appear to be due to a homologous 9-kDa protein found in these fruits (peach; [Pru p 1]), cherry, apricot, and plum) and Brazil nuts (Ber e 1) (14,15). Patients with OAS generally can ingest these foods in the cooked form without difficulty.
Mixed IgE-and Non–IgE-Mediated Disorders
Allergic eosinophilic esophagitis, gastritis, and gastroenterocolitis are characterized by various infiltrations of the esophagus, stomach, and/or intestinal walls with eosinophils, an absence of vasculitis, and peripheral eosinophilia in approximately 50% of patients. The eosinophilic infiltrate may involve the mucosal, muscular, and/or serosal layers of the stomach or small intestine, and clinical symptoms correlate with the extent of eosinophilic infiltration of the bowel wall (16–19). Eosinophilic infiltration of the muscular layer leads to thickening and rigidity-provoking symptoms of obstruction, whereas infiltration of the serosa results in ascites containing eosinophils. Although peripheral blood T cells from these patients have been shown to secrete excessive amounts of interleukin (IL)-4 and IL-5 compared with normal controls (20), the underlying immunopathogenesis of these disorders is poorly understood.
Allergic eosinophilic esophagitis, it is agreed, is seen most frequently during infancy through adolescence, and involves chronic esophagitis with or without reflux, as demonstrated by a 24-hour pH probe study dysphagia, intermittent emesis, food refusal, abdominal pain, irritability, sleep disturbance, failure to respond to conventional reflux medication, and, occasionally, esophageal strictures (Table 4). Total serum IgE is normal or slightly elevated, and peripheral eosinophilia is uncommon. Esophageal biopsy reveals infiltration of the mucosa and submucosa with eosinophils, papillary elongation, and basal zone hyperplasia. Older patients frequently are reactive to several foods, presumably on a non–IgE-mediated basis, because skin test and/or RAST results are frequently negative.
Young infants often respond to extensively hydrolyzed infant formulas, but older patients typically require amino acid–derived formulas for resolution of symptoms (21). Symptoms also respond to systemic steroids (e.g., 2 mg/kg per day) but generally return when steroids are tapered. Symptoms due to one or two foods (e.g., milk) frequently resolve in 1 to 2 years. One study of children less than 1 year of age with gastroesophageal reflux found that 40% of infants studied had cow's milk hypersensitivity and a characteristic phasic pattern on 24-hour pH probe studies (22).
Allergic eosinophilic gastritis, it is agreed, is more common between infancy and adolescence, with symptoms including postprandial vomiting, abdominal pain, anorexia, early satiety, hematemesis, failure to thrive, gastric outlet obstruction (rarely, pyloric stenosis ), and gastric bleeding (Table 5). Approximately half these patients have atopic features, elevated serum IgE levels, and peripheral eosinophilia. Gastric biopsy reveals marked infiltration of the mucosa and submucosa with eosinophils, especially in the gastric antrum. Treatment and outcome are similar to that seen in allergic eosinophilic esophagitis, except that long-term outcome is more guarded in older patients.
Allergic eosinophilic gastroenterocolitis, it is agreed, may occur at any age, with symptoms similar to esophagitis and/or gastritis (Table 6). Weight loss or failure to thrive is a hallmark of this disorder. Up to 70% of patients with this allergic eosinophilic disorder are atopic, and food-induced IgE-mediated reactions have been implicated in a subset of patients. Hypogammaglobulinemia and generalized edema secondary to hypoalbuminemia may occur in some infants with marked protein-losing enteropathy, often in the presence of minimal gastrointestinal symptoms such as occasional vomiting and diarrhea (24). Approximately half of patients are believed to respond to allergen-elimination diets, but resolution of symptoms often requires 3 to 8 weeks after the elimination of responsible food allergens (frequently only a few foods). Biopsy of the esophagus and duodenum reveals infiltration of the mucosa and submucosa with eosinophils, and biopsy of the colon may reveal eosinophils and crypt abscesses.
Similar to the other allergic eosinophilic disorders, allergic eosinophilic gastroenterocolitis responds to elemental diets or systemic steroids. The natural history of this disorder is typically prolonged.
By consensus, dietary protein enterocolitis is a disorder most frequently seen in the first several months of life in which infants show irritability, with protracted vomiting and diarrhea, not infrequently resulting in dehydration (25,26) (Table 7). Vomiting generally occurs 1 to 3 hours after feeding, and diarrhea 5 to 8 hours after feeding. Continued exposure to the food allergen may result in bloody diarrhea, anemia, abdominal distension, and failure to thrive. Symptoms are most commonly provoked by cow's milk or soy protein–based formulas but occasionally result from food proteins passed in maternal breast milk. A similar enterocolitis syndrome has been reported in older infants and children due to egg, wheat, rice, oat, peanut, nuts, chicken, turkey, and fish sensitivity (27). Hypotension occurs in approximately 15% of cases after allergen ingestion (27,28). In adults, shellfish (e.g., shrimp, crab, and lobster) sensitivity may provoke a similar syndrome with severe nausea, abdominal cramps, and protracted vomiting. Stools often contain occult blood, polymorphonuclear neutrophils (PMNs) and eosinophils, and Charcot–Leyden crystals. Prick skin test responses to the suspected foods are negative. Jejunal biopsies classically reveal flattened villi, edema, and increased numbers of lymphocytes, eosinophils, and mast cells. Increased numbers of IgM-and IgA-containing plasma cells are seen in the jejunal mucosa (29). Although the immunopathogenic mechanism of this syndrome remains to be elucidated, studies suggest that food antigen-induced secretion of tumor necrosis factor (TNF)-α from local mononuclear cells may be responsible for the diarrhea and hypotension (30).
Dietary protein proctitis, it is agreed, typically occurs in the first few months of life, with blood-streaked stools in otherwise healthy-looking infants (Table 8). Up to 60% of cases are seen in breast-fed infants (31–33), with the remainder largely in infants fed cow's milk– or soy protein–based formula. Blood loss is typically modest but occasionally produces anemia. Mild hypoalbuminemia and peripheral eosinophilia rarely occur. Bowel lesions generally are confined to the distal large bowel and endoscopy reveals linear erosions and mucosal edema. Biopsy reveals infiltration of eosinophils (>20 eosinophils; 140 × high-power field) and lymphoid nodular hyperplasia in approximately 20% of patients. In severe lesions with crypt destruction, PMNs are also prominent (34). Elimination of the responsible food allergen usually results in resolution of gross bleeding within 72 hours, whereas complete resolution of occult bleeding may take a few weeks. Some children have been found to be sensitive to extensively hydrolyzed infant formulas (e.g., Alimentum, Ross Laboratories; Nutramigen, Mead Johnson, Evansville, IN, U.S.A.) the usual form of therapy in this disorder, and require the use of amino acid–derived formulas (e.g., EleCare, Ross Laboratories, Columbus, OH, U.S.A.; or Neocate, Scientific Hospital Supplies, Gaitherburg, MD, U.S.A.) for resolution of symptoms (35). Most patients are believed to “outgrow” their hypersensitivity in 1 year.
Dietary protein enteropathy, it is agreed, often appears in the first several months of life with diarrhea (not infrequently steatorrhea) and poor weight gain (36). Symptoms include protracted diarrhea, vomiting in up to two thirds of patients, failure to thrive, abdominal distension, early satiety, and malabsorption. (Table 9). Anemia, edema, and hypoproteinemia are occasionally seen. Serum IgE, food-specific IgE, and peripheral eosinophil counts are not abnormal. Cow's milk sensitivity is the most frequent cause of this syndrome in young infants, but it also has been associated with sensitivity to soy, egg, wheat, rice, chicken, and fish in older children. Biopsy reveals patchy villous atrophy with increased crypt length and intraepithelial lymphocytes with few eosinophils. In cow's milk–induced enteropathy, serum IgA and IgG antibodies to cow's milk proteins are elevated. Elimination of the responsible food allergen leads to clearing of symptoms in 3 to 21 days. Most patients are believed to outgrow their hypersensitivity in 2 to 3 years.
Celiac disease, it is recognized, is well described. It is a dietary protein enteropathy characterized by a more extensive loss of absorptive villi and hyperplasia of the crypts leading to malabsorption, chronic diarrhea, steatorrhea, abdominal distension, flatulence, and weight loss or failure to thrive (Table 10). Oral ulcers and other extraintestinal symptoms secondary to malabsorption are not uncommon. Patients with celiac disease are sensitive to gliadin, the alcohol-soluble portion of gluten found in wheat, rye, barley, and possibly oat. Celiac disease is associated with the HLA-DQ2 (and DQ8) haplotype, and approximately 90% of patients with celiac disease ingesting gliadin possess IgA anti-gliadin and anti-endomysium antibodies (37). In addition, celiac disease is associated with dermatitis herpetiformis, diabetes mellitus, thyroid disease, Down syndrome, and IgA deficiency. Histology typically reveals total villous atrophy and extensive cellular infiltrate. The prevalence of celiac disease has been reported in between 1 in 3700 and 1 in 300 people (38), and a recent study among normal blood donors in the United States revealed IgA anti-endomysium antibodies in 1 in 250 subjects (39). Recent studies suggest that the severity of celiac disease varies from a debilitating malabsorption syndrome to a “silent,” subclinical disorder. Chronic ingestion of gluten-containing grains in patients with celiac disease is associated with increased risk of malignancy, especially T-cell lymphoma (40).
Histopathologic studies show that lymphocytes, predominantly of the CD8+ cytotoxic-suppressor phenotype, are prominent in the intraepithelial space, and γ/δ-T cells are increased in the jejunal mucosa and the peripheral blood (41). Recent evidence suggests that celiac disease is associated with increased mucosal activity of tissue transglutaminase (tTGase) on specific protein-bound glutamine (42). Gliadin is one of the few substrates for tTGase, which deamidates specific glutamines within gliadin, creating epitopes that bind efficiently to DQ2 gut-derived T cells (43). tTGase has been shown to be the target of the anti-endomysium antibodies, and antibodies to tTGase may supplement anti-endomysium antibodies as a diagnostic tool. It is now believed that generation of unique gliadin epitopes by tTGase enzymatic modification leads to activation of DQ2 (and DQ8)-restricted T cells, a breakdown in tolerance, and initiation of an autoimmune process that generates inflammation in the gut. Elimination of gliadin from the diet results in downregulation of the T-cell–induced inflammatory process.
Immunologic reactions to foods in the gastrointestinal tract include a continuum of gastrointestinal symptoms. The best-fit method of classification of immunologic gastrointestinal reactions to foods and other dietary products in infants and children is one based on symptom complex and anatomical regions affected (mouth, esophagus, stomach, small intestine, large intestine, rectum), as was used in making the above classification. The characteristic or exclusionary criteria separating specific allergic or allergic-like hypersensitivity reactions from other forms of gastrointestinal disease at each anatomical level include characteristic symptoms and signs, usual age of onset, immune studies (skin test/in vitro test) or biopsy findings, and a response or absence of response to food challenge, food exclusion, specific medications, or other therapies (see tables).
After reviewing the literature and personal experiences of many investigators, participants in the workshop were able to agree on a working classification system for gastrointestinal food allergies. This document is meant to provide a framework for communication and further study and should be viewed as a work in progress. In our attempt to generate this classification system for gastrointestinal allergies, the paucity of hard data in this area was very evident and brought home the need for further research. It is the hope of all participants at the 1998 consensus workshop that a second larger workshop can be held in the future to reassess the utility of the proposed classification system and to generate a more accurate synopsis based on additional years of directed research.
1. Sampson H. Food Allergies. In: Sleisenger M, Fordtran J, Scharschmidt B, Feldman M, eds. Gastrointestinal Disease: Pathophysiology/Diagnosis/Management. Philadelphia: W.B. Saunders, 1993: 1233–40.
2. Rowe AH. Roentgen studies of patients with gastro-intestinal food allergy. JAMA 1933; 100:394–400.
3. Fries JH, Zizmor J. Roetgen studies of children with alimentary disturbances due to food allergy. Am J Dis Child 1937; 54:1239–51.
4. Pollard H, Stuart G. Experimental reproduction of gastric allergy in human beings with controlled observations on the mucosa. J Allergy 1942; 13:467–73.
5. Reimann HJ, Lewin J. Gastric mucosal reactions in patients with food allergy. Am J Gastroenterol 1988; 83:1212–1219.
6. DuMont GCL, Beach RC, Menzies IS. Gastrointestinal permeability in food-allergic eczematous children. Clin Allergy 1984; 14:55–9.
7. Flick J, Sampson H, Perman J. Intestinal permeability to carbohydrates in children with atopic dermatitis and food hypersensitivity (abstract). Pediatr Res 1988; 23:303A.
8. Bircher AJ, Van Melle G, Haller E, Curty B, Frei PC. IgE to food allergens are highly prevalent in patients allergic to pollens with and without symptoms of food allergy. Clin Exp Allergy 1994; 24:367–74.
9. Amlot PL, Kemeny DM, Zachary C, Parkes P, Lessof MH. Oral allergy syndrome (OAS): Symptoms of IgE-mediated hypersensitivity to foods. Clin Allergy 1987; 17:33–42.
10. Ortolani C, Ispano M, Pastorello EA, Ansaloni R, Magri GC. Comparison of results of skin prick tests (with fresh foods and commercial food extracts) and RAST in 100 patients with oral allergy syndrome. J Allergy Clin Immunol 1989; 83:683–90.
11. Anderson L, Dreyfuss E, Logan J, et al. Melon and banana sensitivity coincident with ragweed pollinosis. J Allergy Clin Immunol 1970; 45:310–19.
12. Andersen K, Lowenstein H. An investigation of the possible immunological relationship between allergen extracts from birch pollen, hazelnut, potato, and apple. Contact Dermatitis 1970; 4:73–78.
13. Dreborg S, Foucard T. Allergy to apple, carrot, and potato in children with birch-pollen allergy. Allergy 1983; 38:167–72.
14. Pastorello E, Ortolani C, Farioli L, et al. Allergenic cross-reactivity among peach, apricot, plum, and cherry in patients with oral allergy syndrome: an in vivo and in vitro study. J Allergy Clin Immunol 1994; 94:699–707.
15. Pastorello EA, Incorvaia C, Pravetonni V, et al. New allergens in fruits and vegetables. Allergy 1998; 53:48–51.
16. Katz A, Goldman H, Grand R. Gastric mucosal biopsy in eosinophilic (allergic) gastroenteritis. Gastroenterology 1977; 73:705–9.
17. Lee C, Changchien C, Chen P, et al. Eosinophilic gastroenteritis: 10 years experience. Am J Gastroenterol 1993; 88:70–4.
18. Min K-U, Metcalfe DD. Eosinophilic gastroenteritis. Immunol Allergy Clin North Am 1991; 11:799–813.
19. Moon A, Kleinman R. Allergic gastroenteropathy in children. Ann Allergy Asthma Immunol 1995; 74:5–12.
20. Jaffe J, James S, Mullins G, Braun–Elwert L, Lubensky I, Metcalfe DD. Evidence for an abnormal profile of interleukin-4 (IL-4), IL-5, and gamma interferon in peripheral blood T cells from patients with allergic eosinophilic gastroenteritis. J Clin Immunol 1994; 14:299–309.
21. Kelly KJ, Lazenby AJ, Rowe PC, Yardley JH, Perman JA, Sampson HA. Eosinophilic esophagitis attributed to gastroesophageal reflux: Improvement with an amino-acid based formula. Gastroenterology 1995; 109:1503–12.
22. Iacono G, Carroccio A, Cavataio F, et al. Gastroesophageal reflux and cow's milk allergy in infants: A prospective study. J Allergy Clin Immunol 1996; 97:822–7.
23. Snyder JD, Rosenblum N, Wershil B, Goldman M, Winter HS. Pyloric stenosis and eosinophilic gastroenteritis in infants. J Pediatr Gastroenterol Nutr 1987; 6:543–7.
24. Waldman T, Wochner R, Laster R, et al. Allergic gastroenteropathy. A cause of excessive gastrointestinal protein loss. N Engl J Med 1967; 276:761–9.
25. Powell GK. Enterocolitis in low-birth-weight infants associated with milk and soy protein intolerance. J Pediatr 1976; 88:840–4.
26. Powell GK. Milk-and soy-induced enterocolitis of infancy. J Pediatr 1978; 93:553–60.
27. Sicherer SH, Eigenmann PA, Sampson HA. Clinical features of food protein-induced enterocolitis syndrome. J Pediatr
28. Goldman AS, Anderson DW, Sellers WA, Saperstein A, Kniker WT, Halpern SR. Milk allergy: I. Oral challenge with milk and isolated milk proteins in allergic children. Pediatrics 1963; 32:425–43.
29. Perkkio M, Savilahti E, Kuitunen P. Morphometric and immunohistochemical study of jejunal biopsies from children with intestinal soy allergy. Eur J Pediatr 1981; 137:63–9.
30. Heyman M, Darmon N, Dupont C, et al. Mononuclear cells from infants allergic to cow's milk secrete tumor necrosis factor alpha, altering intestinal function. Gastroenterology 1994; 106:1514–23.
31. Lake AM, Whitington PF, Hamilton SR. Dietary protein-induced colitis in breast-fed infants. J Pediatr 1982; 101:906–10.
32. Machida H, Smith A, Gall D, Trevenen C, Scott RB. Allergic colitis in infancy: Clinical and pathologic aspects. J Pediatr Gastroenterol Nutr 1994; 19:22–6.
33. Odze R, Wershil B, Leichtner A. Allergic colitis in infants. J Pediatr 1995; 126:163–70.
34. Goldman H, Proujanksy R. Allergic proctitis and gastroenteritis in children. Am J Surg Pathol 1986; 10:75–86.
35. Vanderhoof JA, Murray ND, Kaufman SS, et al. Intolerance to protein hydrolysate infant formulas: An underrecognized cause of gastrointestinal symptoms in infants. J Pediatr 1997; 131:741–744.
36. Kosnai I, Kuitunen P, Savilahti E, et al. Mast cells and eosinophils in the jejunal mucosa of patients with intestinal cow's milk allergy and celiac disease of childhood. J Pediatr Gastroenterol Nutr 1984; 3:368–74.
37. Grodzinsky E, Jansson G, Skogh T, Stenhammar L, Falth-Magnusson K. Anti-endomysium and anti-gliadin antibodies as serological markers for coeliac disease in childhood: A clinical study to develop a practical routine. Acta Paediatr 1995; 84:294–8.
38. Cavell B, Stenhammar L, Ascher H, et al. Increasing incidence of coeliac disease in Sweden: results of a national study. Acta Paediatr 1992; 81:589–592.
39. Not T, Horvath K, Hill ID, Partanen J, Hammed A, Magazzu G, Fasano A. Celiac disease risk in the USA: High prevalence of antiendomysium antibodies in healthy blood donors. Scand J Gastroenterol 1998; 33:494–8.
40. Holmes G, Prior P, Lane M, et al. Malignancy in coeliac disease: Effect of a gluten-free diet. Gut 1989; 30:333–8.
41. Klemola T, Tarkkanen J, Ormala T, Saxen H, Savilahti E. Peripheral gamma/delta cell receptor-bearing lymphocytes are increased in children with celiac disease. J Pediatr Gastroenterol Nutr 1994; 18:435–9.
42. Molberg O, McAdam SN, Korner R, et al. Tissue transglutaminase selectively modifies gliadin peptides that are recognized by gut-derived T cells in celiac disease. Nat Med 1998; 4:713–7.
43. Sjostrom H, Lunkin KE, Molberg O, et al. Identification of a gliadin T-cell epitope in coeliac disease: General importance of gliadin deamidation for intestinal T-cell recognition. Scand J Immunol 1998; 48:111–5.