Allergic Proctocolitis (AP), also known as allergic colitis, dietary protein-induced proctocolitis of infancy, food-induced or benign eosinophilicproctocolitis, or Milk/Soy Protein Intolerance, is an often-overlooked manifestation of dietary protein allergy seen predominantly in breastfed infants.
Cow’s milk protein is considered to be the offending antigen in as many as 50% to 65% of cases. Soy protein is also frequently implicated, being the second most common antigen to cause AP. It is estimated that,30% of patients reacting to cow’s milk concomitantly react to soy protein allergens. Other antigens including egg, corn, cereal grains or multiple foods account for the remainder.
Prospective data indicate that approximately 0.5% to 1% of exclusively breastfed infants develop allergic reactions to cow’s milk proteins excreted in the mother’s milk. While infants fed human milk appear to have a lower incidence of allergic reactions to cow’s milk protein than those fed cow’s milk based formula, more than 50% of infants with AP in published reports were exclusively breast fed.Up to 10 % percent of AP cases have a family history of the same.
Perinatal risk factors such as prematurity, maternal health, and timing of introduction of breast milk or formula appear to play no role in AP.
Allergic Proctocolitis appears in the first 2 months of life, with a range of 1 day to 5 months. Bleeding rarely occurs in the first week of life. The mean age at diagnosis is approximately 60 days.
The pathophysiology of GI related food allergy reactions affecting infants and young children is complex. These reactions can be IgE mediated, partially IgE mediated or non IgE mediated. 50% of children affected by food allergies in the first 2 years of life have non-IgE-mediated reactions. A brief review of food allergy pathogenesis and clinical presentation of the various food allergic conditions that can affect the GI tract in infants and toddlers can be found below. Allergic proctocolitis is a non IgE mediated disorder, and can be differentiated from other food allergic conditions by its presentation and more benign clinical course.
Immune Mediated Gastrointestinal Disorders
- Immediate gastrointestinal hypersensitivity
- Oral Allergy Syndrome
- Eosinophilic Gastroenteritis
- Eosinophilic Esophagitis
- Eosinophilic Gastritis
- Allergic Proctocolitis
- Dietary (Food) Protein-Induced Enteropathy
- Dietary (Food) Protein-Induced Enterocolitis Syndrome (FPIES)
- Celiac disease
The GI tract is the largest immunologic organ in the body. It constantly is exposed to antigens from intestinal bacteria and ingested foods. All individuals recognize these antigens as foreign (non- self). A normal response to these antigens would be the development of immune tolerance, an antigen specific suppression of cellular and humoral immune response. Failure to develop immune tolerance will lead to development of food allergy and its unique pathologic immune response that characterizes the different food allergic conditions.
Food allergy is believed to develop in the post-partum period. There are numerous factors that can make it more likely that food allergy develops. First there is a group of foods that account for the majority of all foods reactions, and include milk, egg, peanut, soy, wheat, tree nuts, fish and shellfish. These foods contain allergens with the following characteristics; they are abundant in the food supply, low molecular weight , water soluble, resist thermal and chemical denaturation, contain linear epitopes, are heavily glycosylated (enhances interaction with intestinal dendritic cells). These features allow for potentially allergenic proteins to interact with the immune system. Host feature that may promote development of food allergy include the skewing of the infant immune system towards anallergy promoting, TH2 profile, improper colonization of GI tract with bacteria and the immaturity of the infant gut barrier allowing for increased intestinal permeability. The timing and dose of allergen also appears to be a critical factor in whether tolerance or allergy develops. The critical window for introduction of foods is trying to be determined.Interaction of antigens, intestinal mucosal cells, dendritic cell and intestinal milieu are all critical in whether tolerance or food allergy develop. For more in depth review of food allergy pathogenesis see: Mechanisms of immune tolerance relevant to food allergy, J ALLERGY CLIN IMMUNOL VOLUME 127, NUMBER 3 p.576-584
For an in-depth discussion of the mechanisms of oral tolerance to food protein, read this article.
IgE mediated food reactions result from isotype specific recognition of food antigens by IgE molecules that are bound to Mast Cells and basophils. Cross linking of IgE molecules on these cells leads to degranulation, with immediate release of preformed mediators Subsequent generation of lipid mediators (leukotrienes, prostaglandins, PAF, etc.) will occur which can further expand the reaction. The effect of these mediators on various tissues, lead to the characteristic clinical picture of IgE mediated food reaction. In the GI tract, symptoms can consist of nausea, abdominal cramping, vomiting and diarrhea. These symptoms occur abruptly after ingestion of offending allergen, usually within minutes. They are usually associated with other classic, systemic IgE mediated symptoms including hives, angioedema, respiratory and cardiovascular symptoms and at its most severe, potentially life-threatening systemic anaphylaxis. In food associated cases of anaphylaxis skin symptoms were often absent. Patients with IgE mediated reactions often have positive family history of atopy and may have other associated IgE mediated allergic conditions including atopic dermatitis, allergic rhinitis and asthma.Treatment consists of absolute food avoidance and being prepared to treat allergic reactions with epinephrine and antihistamines if needed.
Eosinophilic gasteroenteropathies can occur anywhere from the esophagus to the colon, including the bile ducts. The symptoms and presentation will depend on the portion of the GI tract involved and whether the inflammation is mucosal or subserosal. The most common eosinophilic condition occurring in infants and young children is eosinophilic esophagitis (EoE). Eosinophilic gasteroeneritis can occur in this age group but more commonly has a latter presentation.EoE is characterized by diffuse eosinophilic inflatration (> 15/ HPF) of the esophagus. In children the condition is usually do to food allergy as withdrawal of appropriate allergens can lead to resolution of the disease. There is a familial tendency, and disease risk genes that have been shown to be involved include Eotaxin-3, TSLP, and TGFB1, which along with IL-5 are involved in the recruitment of eosinophils into the esophagus. In infants and children, presentation usually involves gagging, choking, reflux symptoms, irritability, feeding refusal and poor growth. Diagnosis is made by biopsy showing excessive numbers of eosinophils. Skin testing may be helpful in identifying triggering allergens, although some allergen do not lead to IgE sensitization and cause inflammation via activation of T cell reactive clones. Some investigators have had success using patch testing to identify these allergens, although this is not a standardized technique. Treatment consists of food avoidance and treatment of eosinophilic inflammation with topical steroid therapy.
Food protein-induced enterocolitis syndrome(FPIES) is a more severe form of T cell mediated food allergy, due in part to elaboration of the cytokine tumor necrosis factor from responsive T cells. Children normally present before 9 months, peak incidence is from 1 week to 3 months. Presentation can take one of two forms. First it can present as a sudden, severe, profuse and projectile vomiting, classically occurring 2 hours after exposure to the offending allergen. These children often present to ER appearing quite ill, possibly in shock and lethargic due to dehydration from the profuse vomiting, third spacing of fluid and release of inflammatory mediators. Alternatively, if the offending allergen is chronically in the diet, FPIES can present with severe, chronic gastrointestinal symptoms including vomiting, diarrhea and malabsorption, leading to failure to thrive, poor growth, anemia and hypoproteinemia. Milk and soy are the major allergens but grains, vegetables and poultry have been described as well. Unlike AP, breast feeding is rarely a cause. Treatment consists of avoidance of offending food allergens. The majority, but not all, outgrow by age 3.
Celiac disease is an immune mediated inflammation of the small intestine. Pathologically mucosal inflammation, crypt abscesses and villous atrophy are noted in susceptible patients after ingestion of gluten found in wheat, barley, rye and to a lesser degree oat. The incidence is as high as 0.5 to 1 per cent of the population and occurs in association with HLA DQ2 / DQ8 gene locus, thus tends to run in families. Classic celiac disease occurs at 6 to 24 months with introduction of gluten into the diet. Symptoms include chronic diarrhea, anorexia, abdominal distention, pain and often vomiting. Failure to thrive and malnutrition can occur secondary to malabsorption. A myriad of systemic symptoms can develop as a direct result of the autoimmune process or secondary to malabsorption. Diagnosis can be made by biopsy showing characteristic finding or by serologic testing showing the presence of anti-transglutaminase IGA antibodies. Epidemiological studies have suggested that slow introduction of gluten and continuation of breast feeding during introduction may lessen chance of developing the disease. Treatment is avoidance of gluten in the diet.
The characteristic presentation of AP is discussed below. A T cell mediated cause of AP is suspected. Biopsies of colonic mucosa, have shown increased numbers on intraepithelial CD3+ T cells. These T cells have Tcell receptor repertoire which is consistent with exposure to intraluminal antigens. Eotaxin has been shown to be highly expressed in mucosal biopsies, The cellular source of this elevation has not been determined, but most likely accounts for the eosinophilia seen on colonic biopsies in patients with AP. Endoscopic examination reveals patchy erythema, loss of vascularity and lymphonodular hyperplasia,.Histologic examination reveals preservation of overall architecture of colonic mucosa. There are focal aggregates of eosinophils in the lamina propria, crypt epithelium and musculairis mucosa.
The primary clinical presentation of Allergic Proctocolitis is mucousy, loose stools with spots or streaks of bright red blood in an otherwise healthy infant. The mean age at presentation is approximately 60 days, with a range of 1 day to 6 months. The onset of bleeding is gradual, initially erratic over several days then progressing to streaks of blood in most stools. Increased frequency of bowel movements is possible. Systemic features are absent, except for occasional reports of apparent pain on defecation. Excessive fussiness, regular bouts of inconsolable crying, and interrupted sleep patterns are frequently reported anecdotally, but less mentioned in the literature
Vomiting is absent. Weight gain and growth are normal. The physical exam reveals an otherwise healthy infant. Abdominal examination results are benign. No anal fissures or perianal dermatitis are found to account for the bleeding.
Blood loss is typically modest, but mild anemia and/or hypoalbuminemia may result. Increased eosinophils in fecal smear and mild elevation of peripheral eosinophil count can be seen,but are not diagnostic of the disease. Markers of atopy such as atopic dermatitis, family history of atopy or presence of IgE mediated disease are similar in frequency to the general population. Features of IgE-mediated reactions such as anaphylaxis and hives are lacking.
Signs and symptoms of malabsorption, significant anemia, abdominal distention, severe diarrhea, vomiting, failure to thrive, or impaired growth are absent in AP and should lead to consideration of other conditions.
Bloody, mucousy stools in a neonate or infant understandably are of great concern to parents and can be a reason for serious concern.If bleeding has been mild or self – limited, it is reasonable to defer extensive workup. In patients with mild to moderate illness, differential diagnosis can include viral or gastrointestinal infections, anal fissure, perianal dermatitis, coagulation disorders and vitamin K deficiency. In severe illness, consider, other eosinophilic gastrointestinal disorders, necrotizing enterocolitis, volvulus, Hirschprung’s disease, intussusception, sepsis, inflammatory bowel disease, immunodeficiency and malignancy.
Diagnosis of Allergic Proctocolitis begins with a thorough history and physical examination. Table 2 reviews signs and symptoms to consider in the diagnosis of Allergic Proctocolitis.
Diagnosis of Allergic Proctocolitis
Signs and Symptoms to Consider
Blood-streaked stools, with or without:
- Exclusive breastfeeding
- Mild-to-moderate diarrhea
- Mucousy stools
- Excessive crying/fussiness (colic)
- Sleep disturbances
- Apparent pain on defecation
- Good general condition
- 1 week to 5 month old infant
- Mild Anemia
- Mild, Peripheral eosinophilia
- Family history of AP
- Failure to thrive
- Impaired growth
- Signs and symptoms of malabsorption
- Abdominal distention
- Severe diarrhea
- Significant anemia
- Anal fissures
- Perianal dermatitis
- Ova, parasites, or Clostridium difficil toxin in stool samples
The National Institute of Allergy and Infectious Diseases 2010 expert panel report, Guidelines for the Diagnosis and Management of Food Allergy in the United States, “recommends using the medical history, resolution of symptoms when the causative food is eliminated from the diet, and recurrence of symptoms following an oral food challenge to diagnose AP.”The American Gastroenterological Association concurs, stating that resolution of symptoms after withdrawal of the presumed allergen is considered sufficient to make the diagnosis.
Lake’s review of 95 exclusively breast fed infants with AP in Journal of Pediatric Gasteroenterology and Nutrition In January, 2000, showed elimination of a single dietary protein in mother’s diet led to resolution. Five mothers needed to eliminate 2 or more foods and eleven patients needed to be switched to a protein hydrolysate despite multiple dietary interventions. With appropriate intervention, the vast majority of mothers did not need to stop nursing sooner than they had planned. Infants of mothers, who continued nursing their AP child with persistent rectal bleeding, either to being unable to identify the offending protein or to non-compliance with the avoidance diet, had persistent symptoms and were more likely to become anemic.
Empiric food avoidance in this condition correctly identifies and treats two thirds of patients. Alternatively, this implies that in one third of cases alteration of the maternal diet or use of a more expensive formula may be done unnecessailry.
A food challenge can be conducted to confirm the diagnosis at the discretion of the health care practitioner. Challenges are usually performed in an open fashion (not double blinded), with the asymptomatic patient redeveloping symptoms within 48 to 72 hours of reexposure to the offending allergen.If an IgE-mediated reaction or FPIES is considered, food challenges must be conducted under close supervision in a facility equipped to handle severe allergic reactions.
Food allergy is the major cause of rectal bleeding due to colitis in infants. Allergic Proctocolitis should always be part of the differential diagnosis for an infant presenting with blood-streaked stools. An algorithm for diagnosis of infants with haematochezia and blood streaked stool, proposed by Pumberg et al.
No diagnostic tests are usually necessary if clinical presentation supports the diagnosis.
Tests for anemia and hypoalbuminemia may be warranted if bleeding has been ongoing.
A CBC may show mild eosinophilia but is not diagnostic. Absence of peripheral eosinophilia, does not rule out a diagnosis of AP.
Tests for total and antigen-specific IgE concentrations are not helpful in diagnosing non-IgE-mediated reactions such as AP. Patch testing has been tried with some success in some patients who have not responded to a removal of suspected antigen from diet. Again this is not a standardized technique. Researchers are investigating tests that may help in the diagnosis of non-IgE-mediated allergies in the future. In their 2011 review, Milk and Soy Allergy, Kattan et al. discuss protein microarrays using purified natural cow’s milk-allergen components as a possible diagnostic test currently under investigation.
Stool cultures may be performed to rule out pathogenic bacteria, parasitic ova and the presence of Clostridium difficile toxin.
Radiographic investigations can be performed if warranted to rule out other conditions, such as, necrotizing enterocolitis and Hirschsprung disease.
Testing for coagulation disorders is not justified in cases of mild bleeding.
Endoscopy and biopsy are not necessary for diagnosis, unless the clinical picture is not clear.Endoscopy has been used in research to elucidate the condition. Focal erythema and friability are the most commonly seen findings in Allergic Proctocolitis. Patchy focal or diffuse mucosal edema, loss of vascularity, and small erosions are also noted. Area of involvement is predominantly the sigmoid colon and rectum. Kattan et al. described lymphonodular hyperplasia in up to 40%. In patients with more chronic manifestations at diagnosis, inflammatory features of ulceration in the rectum can be seen.
Histological examination typically shows high numbers of eosinophils and plasma cell infiltration of the epithelium, lamina propria, and muscularis mucosa. Multinucleated giant cells in the submucosa have been reported. Kattan et al. found that the regular structure of the mucosa is preserved, but all specimens showed a striking hyperemia.
To reiterate, removal of suspected food allergen from infant diet, with resolution of symptoms is sufficient for diagnosis.
The treatment for AP is the elimination of the offending protein(s) from the diet. For breastfeeding infants, this means a maternal elimination diet. Since cow’s milk protein is most frequently implicated in AP, mothers should be instructed to eliminate all sources of cow’s milk protein for a minimum of 2 weeks. Patient education information on carrying out a dairy elimination diet is available elsewhere on this site[JUMP to Elim. Diet/patient handouts] In most cases, obvious bleeding will resolve in 72 to 96 hours. An elimination period of up to 4 weeks may be necessary in some cases, especially if bleeding has been prolonged. If improvement is not seen after 2 weeks of a strict dairy-free diet, soy protein should be eliminated as well. Note: Given the frequency of concomitant sensitization to both dairy and soy, many providers choose to recommend elimination of both proteins immediately, with or without a trial re-introduction of soy after symptoms have resolved.
Some infants are more sensitive than others to small amounts of antigen exposure. While recurrence of bleeding has been noted in cow’s milk-sensitive infants after inadvertent maternal intake of only a pat of butter, some infants may tolerate dairy or soy protein that has been heated, fermented, or otherwise broken down. Mother’s should begin with total elimination, including “hidden” sources of all protein fractions. Maintaining nutritional adequacy on such a strict elimination diet is challenging and may be emotionally trying for new mothers. Referral to a Registered Dietitian to guide elimination and assure proper nourishment of the breastfeeding mother is strongly recommended.
Breastfeeding mothers on a diary-elimination diet should be advised to take calcium supplements (1,200mg/day divided into several doses).
If bleeding continues after 2 to 4 weeks of maternal elimination of dairy and soy:
- look for “hidden” dairy and/or soy in the mother’s diet
- consider less frequently implicated proteins, such as egg or corn
If there is no improvement in response to carefully controlled elimination trials of common allergens:
- Consider referral to a pediatric gastroenterologist and/or pediatric allergist.
- Monitor for anemia and hypoalbuminemia.
- Monitor weight gain and growth.
- Consider switching to a hypoallergenic or elemental formula. The vast majority of woman will be able to continue to breast feed.
For infants receiving cow’s milk formula, switching to soy-based formula is often the first step in treatment. If symptoms do not resolve, or soy has already been determined to be a provoking allergen, a switch to an extensively hydrolyzed formula will be necessary. Given the frequency of both dairy and soy being allergens, many providers choose to recommend extensively hydrolyzed formula without a trial period of soy formula. In rare cases, some patient shave reacted to protein hydrolysate formulas and have needed to use, elemental amino acid based formulas.
If symptoms resolve with elimination, the allergen may be reintroduced into the diet if desired to confirm diagnosis. Reintroduction usually provokes recurrence of symptoms within 72 hours. If symptoms recur upon challenge, diagnosis is confirmed, and the offending protein should be eliminated from the mother’s (and infant’s) diet for at least 6 months and until 9 to 12 months of age.
If the elimination diet is successful in controlling symptoms, no other action is required beyond standard care.
When solid foods are introduced, all sources of the offending protein should be avoided. Other foods can be introduced at appropriate times and do not need to be delayed.According to the National Institute of Allergy and Infectious Diseases 2010 expert panel report, Guidelines for the Diagnosis and Management of Food Allergy in the United States, “the introduction of solid foods should not be delayed beyond 4 to 6 months of age. Potentially allergenic foods may be introduced at this time as well.”
Six months after diagnosis, as long as the baby is at least 9 months old, the food can be gradually reintroduced into the diet. Because of the benign nature of this condition and because of the delay in time between reintroduction and symptom development, reintroduction can be done at home, with close communication between family and medical professional. Again, if there is any confusion that an IgE mediated or FPIES syndrome is involved then a medically supervised,in office food challenge should take place.
A small study by Repucci looked at the use of maternal ingestion of pancreatic enzymes to treat cases of stool blood in breast fed infants that did not respond to extensive elimination diets. Bleeding resolved in 3 out of 4 infants with administration of two capsules of Pancrease MT 4® (USP Units: 4,000 lipase/12,000 amylase, 12,000 protease) with meals and one with snacks. Bleeding resolved in the fourth infant with maternal ingestion if three capsules with meals and two with snacks. Elimination diet plus Pancrease MT4® brought about resolution of bleeding and reduction in colicky symptoms in 13 of 16 infants at the breastfeeding clinic of the University of California, Davis Medical Center in Sacramento. It is theorized that the pancreatic enzymes further break down the dietary proteins before they enter the mother’s milk, leaving less intact protein to trigger an allergic reaction in the infant’s intestines. There have been no prospective case-controlled studies of this treatment.
A report in Journal of Pediatrics, March 2010, 397-401., suggested addition of probiotic Lactobacillus to protein hydrolysate formula more rapidly and completely lead to healing of the colon compared to formula alone.
Both of these treatments could be considered, but are not part of standard protocols.
Allergic Proctocolitis resolves by age 1-2 years in most children. Rarely, symptoms persist into later childhood. Of 35 infants followed by Lake et al. for more than 10 years, none developed intestinal disease.
American Gastroenterological Association. AGA Technical Review on the Evaluation of Food Allergy in Gastrointestinal Disorders. Gastroenterology 2001;120:1026-1040.
Sicherer SH. Food Allergy. The Lancet 2002; 360(9334):701-710. Available at:
Pumberger W, Pomberger G, Geissler W. Proctocolitis in breast fed infants: a contribution to differential diagnosis of haematochezia in early childhood. Postgrad Med J 2001;77:252-254.
Kattan JD et al. Milk and Soy Allergy.KPediatrClin North Am 2011; 58(2):407-426.
Vandenplas Y et al. Guidelines for the diagnosis and management of cow’s milk protein allergy in infants. Arch Dis Child 2007;92:902-908.
Brill H. Approach to milk protein allergy in infants. Can Fam Physician 2008; 54(9): 1258–1264.
Lake, AM. Food-Induced Eosinophilic Proctocolitis. Journal of Pediatric Gastroenterology & Nutrition 2000; 30(1):S58-S60.
Academy of Breastfeeding Medicine. ABM Clinical Protocol #24: Allergic Proctocolitis in the Exclusively Breastfed Infant. Breastfeeding Medicine 2001; 6(6):435-440.
Troncone R, Discepolo V. Colon in Food Allergy. Journal of Pediatric Gastroenterology & Nutrition 2009;48:S89-S91.
Liacouras C. Evaluation and Treatment of Non-Ige Mediated Gastrointestinal Food Allergy. Journal of Pediatric Gastroenterology & Nutrition 2004;39:S549.
Pabst O, Mowat AM. Oral tolerance to food protein. Mucosal Immunology 2012; 5: 232–239.
Gibbons T, et al.. Non-IgE-Mediated Cow Milk Allergy Is Linked to Early Childhood Clusters of Commonly Seen Illnesses. ClinPediatr 2012; 51(4):337-344
du Toit G, et al.. Best Practice: Identifying and managing cow’s milk protein allergy. Arch Dis Child EducPract Ed 2010;95(5):134-144.
Stiefel G, Roberts G. How to use serum-specific IgE measurements in diagnosing and monitoring food allergy. Arch Dis Child EducPract Ed 2012;97:29-36.
Center for Clinical Practice, National Institute for Health and Clinical Excellence, British National Health Service. Quick reference guide: Food allergy in children and young people. 2011.
Repucci A. Abstract: Resolution of Stool Blood in Breast-Fed Infants With Maternal Ingestion of Pancreatic Enzymes. Journal of Pediatric Gastroenterology & Nutrition 1999; 29(4):500.
Nocerino A, Protein Intolerance.
Lock RJ, Unsworth DJ. Food allergy: which tests are worth doing and which are not. Ann ClinBiochem 2011;48(4): 300-309.
Lucarelli S et al. Allergic proctocolitis refractory to maternal hypoallergenic diet in exclusively breast-fed infants: a clinical observation. BMC Gastroenterol 2011; 11: 82.
Kilshaw PJ, Cant AJ. The passage of maternal dietary proteins into human breast milk. Int Arch Allergy ApplImmunol 1984;75(1):8-15.
Odze RD, Wershil BK, Leichtner AM, et al.. Allergic colitis in infants. J Pediatr 1995; 126:163-70.
Saarinen KM, Suomalainen H, Savilahti E. Diagnostic value of skin-prick and patch tests and serum eosinophil cationic protein and cow’s milk-specific IgE in infants with cow’s milk allergy. ClinExp Allergy 2001 Mar;31(3):423-9.
Husby S. Food Allergy as Seen by a Paediatric Gastroenterologist. Journal of Pediatric Gastroenterology& Nutrition 2008;(47):S49-S52.
Sampson H. Non-Ige Mediated Food Allergy: New Diagnostic Challenges. Journal of Pediatric Gastroenterology & Nutrition 2004; 39:S549-S550.
Fretzayas A, et al. Thrombocytosis as an overt sign of cow’s milk allergic proctocolitis. AllergolImmunopathol (Madr.). 2011. Doi:10.1016/j.aller.2010.11.001
Boné J, Claver A, Guallar I, Plaza AM. Allergic Proctocolitis, food-induced enterocolitis: immune mechanisms, diagnosis and treatment. AllergolImmunopathol (Madr) 2009;37(1):36-42.
Guidelines for the Diagnosis and Management of Food Allergy in the United States Summary of the NIAID-Sponsored Expert Panel Report.