Sicherer, Scott H.
Food allergy, an immune-mediated disorder, affects 6% to 8% of infants less than 2 years of age (1). The most common dietary proteins responsible for food-allergic reactions in these infants are cow's milk and soybean, with cow's milk allergy (CMA) accounting for approximately 2.5% of the allergic responses (1–4). The clinical manifestations of food allergy include acute reactions (e.g., hives, asthma, and anaphylaxis) and chronic, indolent disease (e.g., failure to thrive, diarrhea, and atopic dermatitis) (5). These disorders may be divided on an immunologic basis into those that are mediated by IgE antibody directed toward the specific causal protein and those that are not IgE mediated. In general, food-allergic disorders affecting only the gut are non–IgE mediated (with the exception of subsets of children with “gastrointestinal anaphylaxis,” eosinophilic gastroenteritis, and malabsorption associated with atopic dermatitis ). Fortunately, 85% to 90% of infants with CMA show tolerance by 3 years of age (2,7,8).
Among the food-allergic disorders affecting the gastrointestinal tract, there are a myriad of clinical manifestations (vomiting, diarrhea, growth failure, abdominal pain) and a broad range of severity and chronicity. Clinical–histopathologic correlation includes a variety of findings with varying degrees and types of inflammation. Some food-allergic disorders affecting the gut have been well delineated (celiac disease ). However, many of these disorders (e.g., enteropathy syndrome, gastroesophageal reflux, eosinophilic gastroenteritis, and others) remain less well defined, may represent a spectrum of disorders, are often “lumped” together in clinical reviews, and probably represent a number of immunopathophysiologic causes. The primary feature common to the food-allergic disorders affecting the gut is that elimination of the offending food protein results in resolution of the clinical symptoms (and normalization of histologic abnormalities), reintroduction of the causal protein results in recurrence of these abnormalities, and no other cause (infection, intolerance) is identifiable (10).
Food protein-induced enterocolitis syndrome (FPIES) describes a symptom complex of profuse vomiting and diarrhea diagnosed in infancy (11–13). Because both the small and large bowel are involved, the term enterocolitis is used. Although the clinical syndrome may represent a more severe form of food protein-induced enteropathy (abdominal symptoms, malabsorption with associated villous blunting (14)), there are particular features including signs of systemic reactivity and implications for long-term clinical management that warrant its discussion as a distinct clinical entity. The purpose of this review is to delineate the characteristic clinical features, diagnosis, and management of FPIES. Furthermore, infantile FPIES will be discussed in relation to other clinical syndromes that share features with it (atypical FPIES). Finally, the place of FPIES as a distinct clinical entity will be discussed.
HISTORICAL BACKGROUND AND DEFINITION OF FPIES
In 1967, Gryboski (15) described a series of 21 children diagnosed with gastrointestinal symptoms of CMA. Most of the children began to experience symptoms during the first 6 weeks of life. Eleven of 21 had growth retardation, and 7 of 21 experienced symptoms of shock after reintroduction of the causal protein. Almost all the infants tolerated cow's milk by 3 years of age. A decade later, Powell (16) described two premature infants with symptoms of recurrent vomiting, bloody diarrhea, and abdominal distension with suspected necrotizing enterocolitis. For these two infants, no bacterial pathogens or other causes of necrotizing enterocolitis could be identified. However, both infants experienced resolution of their symptoms when not ingesting cow's milk or soybean infant formula, and both had recurrence of symptoms (vomiting, bloody diarrhea, and malabsorption) and elevation of their peripheral polymorphonuclear (PMN) leukocyte count when these formulas were reintroduced (casein hydrolysate formula was tolerated).
In a subsequent study, Powell (11) described nine infants with severe, protracted diarrhea and vomiting. The symptoms developed 4 to 27 days after birth (mean, 11 days) while the infants were consuming a cow's milk–based formula. Switching to a soybean-based formula resulted in transient improvement, but symptoms generally recurred in 7 days. Seven of the nine infants were below birth weight, and eight of nine had dehydration. Eight of the infants appeared acutely ill and underwent sepsis evaluations (negative). All infants were noted to have low serum albumin and an elevated PMN leukocyte, count and stools were positive for heme and reducing substances. The hospital course for these infants usually involved improvement after administration of intravenous fluids, followed by recurrence of dramatic symptoms with reintroduction of soy-or cow's milk–based formula, including shock in several patients. Formal follow-up oral challenges were performed with cow's milk and soybean (30–100 ml of formula) at a mean age of 5.5 months. Fourteen of the 18 challenges elicited positive responses. Ten of 14 challenges resulted in vomiting (onset, 1–2.5 hours after ingestion; mean, 2.1 hours), and all experienced diarrhea (onset, 2–10 hours; mean, 5 hours) with blood, leukocytes, eosinophils, and increased carbohydrate in the stool. There was an increase in PMN leukocyte count in all positive challenges, peaking 6 hours after ingestion with a mean increase of 9,900 cells/mm3 (range, 5,500–16,800 cells/mm3). Only gastrointestinal symptoms were reported.
The results of these studies (11,16) led Powell to propose a particular challenge protocol (Table 1) and criteria for a positive oral challenge to diagnose food protein–induced enterocolitis of infancy (Table 2) (12). Powell recognized that other causes of symptoms that could mimic FPIES must be ruled out (infections, inflammatory bowel disease, ischemia, and others) and that normal weight gain and abatement of symptoms could be expected to occur when these infants were ingesting a hydrolyzed formula.
CASE REPORT: AN EXAMPLE OF COW'S MILK–INDUCED ENTEROCOLITIS
The patient was a full-term male infant who was initially fed with a cow's milk–based formula. Between 7 and 14 days of age, progressive vomiting developed, with heme-positive diarrhea, and subsequent lethargy. He was admitted for a full sepsis evaluation (negative), and the symptoms resolved after he underwent intravenous hydration. He was discharged from the hospital with a soybean-based formula that was tolerated in the hospital. Because of mild symptoms of colic, at 8 weeks of age the formula was switched to a casein hydrolysate. Two hours after a first feeding of this formula, protracted vomiting occurred, and the soybean-based formula was therefore resumed. Accidental ingestion of a small amount of cow's milk at 5 months of age resulted in protracted vomiting (1 hour later), diarrhea (6 hours later), and acidemia and cyanosis requiring hospital admission. Milder symptoms occurred after ingestion of a tiny amount of cheese at 11 months of age.
The patient successfully avoided cow's milk from 11 months to 6 years of age. Results of prick skin tests and radioallergosorbent test (RAST) for cow's milk were negative. He was admitted to the hospital, and intravenous access was obtained. He received 1 oz (approximately 0.06 g/kg) of cow's milk protein in a single-blind challenge. He was without symptoms until 90 minutes after the ingestion, when abdominal pain and repetitive vomiting began and continued for 3 hours. At that time the peripheral blood PMN leukocyte count (including immature forms) had increased by 14,713 cells/mm3. There were no skin or respiratory symptoms. The diastolic blood pressure decreased by 10 mmHg. He was treated with intravenous fluids and then intravenous steroids. Symptoms resolved, and he was discharged home. Loose stools were noted the next day.
To summarize the observations of Powell (11,12,16), infantile FPIES is a severe syndrome of vomiting and diarrhea caused by milk and/or soy in infants. Confirmation of the allergy includes a negative search for other causes, improvement when not ingesting the causal protein, and a positive response to oral challenge, as described. Infants with symptoms consistent with severe enterocolitis who fulfilled, or are highly likely to have fulfilled, these criteria are included in many reports of milk or soy allergy of infancy (17–24). Some additional information about the clinical characteristics of these infants has emerged (Table 3). In a review of 17 infants admitted to hospitals with FPIES, Murray and Christie (25) reported six infants with acidemia (mean pH 7.03) and methemoglobinemia. It was hypothesized that this resulted from increased heme oxidation caused by an elevation of nitrites in the intestine. This elevation of nitrites can occur with severe intestinal inflammation, resulting in reduced catalase activity. The association of methemoglobinemia with FPIES was noted in the current case report and in our published experience with FPIES (13).
Several other clinical features of infantile FPIES have emerged from our studies (13) (16 patients) and a study by Burks et al. (24) (43 patients). All the infants studied had negative skin-prick and/or RAST responses to the causal proteins (cow's milk and/or soy). Approximately half the infants reacted to both milk and soy. Sensitivity to milk was lost in 60% and to soy in 25% of the patients after 2 years from the time of initial occurrence. Lastly, while Powell challenged with approximately 0.6 g protein/kg body weight, in children with severe prior reactions as described in the current case report, lower doses may be adequate (0.06–0.6 g/kg)
Powell noted (12) that her strict challenge criteria for infantile FPIES would potentially exclude some patients who might have a similar or identical clinical manifestation or underlying pathophysiology but with differences in particular specific features or severity. Additionally, some patients who fulfill the challenge criteria may have particular features that do not satisfy the general description of infantile FPIES (13) : reactions to foods other than cow's milk or soy, onset of symptoms beyond 9 months of age, and development or presence of IgE antibody directed toward the causal protein. For example, symptoms consistent with FPIES in infants have been described with foods including rice (26), poultry (13,27), fish (28), and pea (13). Similar to the the patient described earlier, some patients with FPIES have experienced reactions to the residual protein in hydrolyzed cow's milk formula (13,29). Onset of FPIES outside of infancy may also occur (13). Lastly, we (13) have noted two patients who had typical infantile FPIES with negative skin-prick and RAST test results in whom positive test results were obtained 1 and 3 years after the diagnosis. In addition, three patients with positive skin-prick responses and RAST results to soybean (ages 7 months, 1 year, 9 years) showed only symptoms consistent with FPIES when orally challenged with soy. All five of these patients who eventually showed or had shown IgE antibody to the causal protein remained sensitive.
DIAGNOSIS AND MANAGEMENT
As noted, the diagnosis rests on clinical and challenge criteria. In practicality, many patients would not undergo a formal challenge as infants because the diagnosis becomes self-evident after elimination of the causal protein. Unfortunately, this has been a hindrance to better studies of this syndrome. For treatment, as noted earlier, most infants do well consuming a casein hydrolysate formula. Because there is a high percentage of patients with sensitivity to both cow's milk and soy, switching directly to a hydrolysate is recommended. For the rare patients reactive to hydrolysate, an amino acid–based formula is appropriate (29). Families must be instructed about the careful avoidance of cow's milk or soy. Lists containing code words indicating cow's milk protein (i.e., casein, whey, natural flavoring, and others), and soy should be provided.
Follow-up challenges should be performed at intervals to determine tolerance (approximately every 18–24 months, depending on the clinical severity). These challenges should be performed under a physician's supervision with emergency medications immediately available, because dramatic reactions, including shock, can occur. Re-evaluation for development of antigen-specific IgE antibody may be helpful (13). Our (13) experience has been that approximately half of positive challenges require treatment (usually intravenous fluids). Because of the presumed pathophysiology, corticosteroids have been administered for severe reactions. The role of epinephrine for treatment is not known, but it should be available for severe cardiovascular reactions.
Unfortunately, because infantile FPIES is a diagnosis that can be made clinically, there are no series in which biopsies have been performed solely in patients fulfilling Powell's criteria. Thus, specific descriptions of the histologic findings are not available, and only assumptions can be drawn by considering descriptions from case reports or series that probably included these patients. The findings from endoscopy and biopsy in FPIES are nonspecific. Colonic specimens in symptomatic patients reveal crypt abscesses and a diffuse inflammatory cell infiltrate with prominent plasma cells, and small bowel specimens reveal edema, acute inflammation and mild villus injury (15,30–33). In some cases, focal erosive gastritis and esophagitis are found, with prominent eosinophilia and villus atrophy (31,34,35).
Van Sickle et al. (36) studied patients fulfilling Powell's strict criteria for FPIES and noted that in vitro stimulation of peripheral blood mononuclear cells with the causal antigen resulted in greater cell proliferation than in children with negative challenges. Also, an increase in serum antigen-specific IgA has been noted in these patients (37). The pathophysiologic ramifications and clinical implications of these findings remain unanswered. In other studies in patients who may have fulfilled the criteria for FPIES, a number of interesting observations have been made. Morphologic studies of the intestine show an increase of IgG-and IgM-containing plasmocytes, as well as an increase of intraepithelial lymphocytes (21). However, the presence of plasmocytes may be the result of a nonspecific immune response caused by an increased passage of antigen through the gut mucosa. The immune response to antigen in the gastrointestinal tract may primarily involve T cells. Cytokines secreted by activated T cells significantly influence the integrity of the mucosal barrier of the gut on exposure to the antigen. Heyman et al. (22) demonstrated that tumor necrosis factor (TNF)-α secreted by circulating milk protein-specific T cells increased intestinal permeability, thus contributing to the influx of antigen into the submucosa with further activation of antigen-specific lymphocytes. Fecal TNF-α was also found in increased concentrations after positive milk challenge in patients with enterocolitis (38). Benlounes et al. (23) showed that significantly lower doses of intact cow's milk protein stimulates TNF-α secretion from peripheral blood mononuclear cells of patients with active intestinal CMA, compared with patients whose sensitivity resolved or with those with skin, rather than intestinal, manifestations of cow's milk hypersensitivity. Interferon (IFN)-γ significantly enhances the action of TNF-α on intestinal epithelial cells, and interleukin (IL)-4 has been determined to have a similar effect on the gut mucosa (39). Conversely, transforming growth factor (TGF)-β1 protects the epithelial barrier of the gut from the penetration of foreign antigens by preventing the detrimental action of IFN-γ(40).
THE SYNDROME IN PERSPECTIVE
As has been described, infantile FPIES appears as a particularly severe gastrointestinal reaction to milk or soy protein. However, FPIES is also part of a clinical spectrum of non–IgE-mediated gastrointestinal food-allergic disorders. It may simply represent a more severe, generalized form of food-associated enteropathy, (14,23,41) proctocolitis, (31,42) or an earlier occurrence of eosinophilic gastroenteritis, (43) or it may be a distinct pathologic entity. The relationship of atypical FPIES to these food-allergic disorders remains to be clarified. It may be that older patients, even adults, with, for example, shellfish-induced vomiting and diarrhea in the absence of specific IgE antibody have a form of FPIES with a pathophysiologic basis similar to that of the infantile form. Clearly, further study (including histologic and immunologic correlates, and long-term clinical follow-up) of this and the other IgE and non–IgE-mediated food-allergic disorders affecting the gut is needed.
1. Bock SA. Prospective appraisal of complaints of adverse reactions to foods in children during the first 3 years of life. Pediatrics 1987; 79:683–8.
2. Host A, Halken S. A prospective study of cow milk allergy in Danish infants during the first 3 years of life. Allergy 1990; 45:587–96.
3. Hide DW, Guyer BM. Cow milk intolerance in Isle of Wight infants. Br J Clin Prac 1983; 37:285–7.
4. Schrander JJP, van den Bogart JPH, Forget PP, Schrander– Stumpel CTRM, Kuijten RH, Kester ADM. Cow's milk protein intolerance in infants under 1 year of age: A prospective epidemiological study. Eur J Pediatr 1993; 152:640–44.
5. Sicherer SH. Manifestations of food allergy: Evaluation and management. Am Fam Phys 1999; 59:415–30.
6. Sampson HA. Food allergy. JAMA 1997; 278:1888–94.
7. Host A, Jacobsen HP, Halken S, Holmenlund D. The natural history of cow's milk protein allergy/intolerance. Eur J Clin Nutr 1995; 49(suppl 1):S13–S8.
8. Bock SA. The natural history of food sensitivity. J Allergy Clin Immunol 1982; 69:173–7.
9. Trier JS. Celiac sprue. N Engl J Med 1991; 325:1709–19.
10. Walker–Smith JA. Diagnostic criteria for gastrointestinal food allergy in childhood. Clin Exp Allergy 1995; 25(suppl 1):20–2.
11. Powell GK. Milk-and soy-induced enterocolitis of infancy. J Pediatr 1978; 93:553–60.
12. Powell G. Food protein-induced enterocolitis of infancy: Differential diagnosis and management. Comp Ther 1986; 12:28–37.
13. Sicherer SH, Eigenmann PA, Sampson HA. Clinical features of food protein-induced enterocolitis syndrome. J Pediatr 1998; 133:214–9.
14. Walker–Smith JA. Cow milk-sensitive enteropathy: Predisposing factors and treatment. J Pediatr 1992; 121:S111–S5.
15. Gryboski J. Gastrointestinal milk allergy in infancy. Pediatrics 1967; 40:354–62.
16. Powell GK. Enterocolitis in low-birth-weight infants associated with milk and soy protein intolerance. J Pediatr 1976; 88:840–4.
17. Hill DJ, Firer MA, Shelton MJ, Hosking CS. Manifestations of milk allergy in infancy: Clinical and immunological findings. J Pediatr 1986; 109:270–6.
18. Fontaine J, Navarro J. Small intestinal biopsy in cow's milk protein allergy in infancy. Arch Dis Child 1975; 50:357–62.
19. Kuitunen P, Visakorpi J, Savilahti E, Pelkonen P. Malabsorption syndrome with cow's milk intolerance: Clinical findings and course in 54 cases. Arch Dis Child 1975; 50:351–6.
20. Hill DJ, Firer MA, Ball G, Hosking CS. Natural history of cows' milk allergy in children: Immunological outcome over 2 years. Clin Exp Allergy 1993; 23:124–31.
21. 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.
22. 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.
23. Benlounes N, Dupont C, Candalh C, et al. The threshold for immune cell reactivity to milk antigens decreases in cow's milk allergy with intestinal symptoms. J Allergy Clin Immunol 1996; 98:781–9.
24. Burks AW, Casteel HB, Fiedorek SC, Williams LW, Pumphrey CL. Prospective oral food challenge study of two soybean protein isolates in patients with possible milk or soy protein enterocolitis. Pediatr Allergy Immunol 1994; 5:40–5.
25. Murray K, Christie D. Dietary protein intolerance in infants with transient methemoglobinemia and diarrhea. J Pediatr 1993; 122:90–2.
26. Borchers SD, Li BUK, Friedman RA, McClung HJ. Rice-induced anaphylactoid reaction (case report). J Pediatr Gastroenterol Nutr 1992; 15:321–4.
27. Vandenplas Y, Edelman R, Sacre L. Chicken-induced anaphylactoid reaction and colitis. J Pediatr Gastroenterol Nutr 1994; 19:240–1.
28. Vitoria JC, Camarero C, Sojo A, Ruiz A, Rodriguez–Soriano T. Enteropathy related to fish, rice and chicken. Arch Dis Child 1982; 57:44–8.
29. Kelso JM, Sampson HA. Food protein-induced enterocolitis to casein hydrolysate formulas. J Allergy Clin Immunol 1993; 92: 909–10.
30. Goldman H, Provjanksy R. Allergic proctitis and gastroenteritis in children. Am J Surg Pathol 1986; 10:75–86.
31. Lake AM. Food protein-induced colitis and gastroenteropathy in infants and children. In: Metcalfe DD, Sampson HA, Simon RA, eds. Food allergy: Adverse reactions to foods and food additives. Boston: Blackwell Scientific Publications, 1997: 277–86.
32. Halpin TC, Byrne WJ, Ament ME. Colitis, persistent diarrhea, and soy protein intolerance. J Pediatr 1977; 91:404–7.
33. Jenkins H, Pincott J, Soothill J, Milla P, Harries J. Food allergy: The major cause of infantile colitis. Arch Dis Child 1984; 59:326–9.
34. Forget PP, Arenda JW. Cow's milk protein allergy and gastroesophageal reflux. Eur J Pediatr 1985; 144:298–300.
35. Coello–Ranurez P, Larrosa–Haro A. Gastrointestinal occult hemorrhage and gastroduodenitis in cow's milk protein intolerance. J Pediatr Gastroenterol Nutr 1984; 3:215–8.
36. Van Sickle GJ, Powell GK, McDonald PJ, Goldblum RM. Milk-and soy protein-induced enterocolitis: Evidence for lymphocyte sensitization to specific food proteins. Gastroenterology 1985; 88:1915–21.
37. McDonald PJ, Goldblum RM, Van Sickle GJ, Powell GK. Food protein-induced enterocolitis: altered antibody response to ingested antigen. Pediatr Res 1984; 18:751–5.
38. Majamaa H, Miettinen A, Laine S, Isolauri E. Intestinal inflammation in children with atopic eczema: Faecal eosinophil cationic protein and tumour necrosis factor-alpha as non-invasive indicators of food allergy. Clin Exp Allergy 1996; 26:181–7.
39. Colgan S, Resnick M, Parkos C, et al. IL-4 directly modulates function of a model human intestinal epithelium. J Immunol 1994; 153:2122–9.
40. Planchon S, Martins C, Guerrant R, Roche J. Regulation of intestinal epithelial barrier function by TGF-beta 1. J Immunol 1994; 153:5730–9.
41. Walker–Smith JA. Food sensitive enteropathies. Clin Gastroenterol 1986; 15:55–69.
42. Lake AM, Whitington PF, Hamilton SR. Dietary protein-induced colitis in breast-fed infants. J Pediatr 1982; 101:906–10.
43. Min K-U, Metcalfe D. Eosinophilic gastroenteritis. Immunol Allergy Clin N Am 1991; 11:799–813.
© 2000 Lippincott Williams & Wilkins, Inc.