Protein Intolerance

Updated: Oct 20, 2017
Author: Agostino Nocerino, MD, PhD; Chief Editor: Carmen Cuffari, MD 



Many food proteins can act as antigens in humans. Cow's milk proteins are most frequently implicated as a cause of food intolerance during infancy. Soybean protein ranks second as an antigen in the first months of life, particularly in infants with primary cow's milk intolerance who are placed on a soy formula. From school age on, egg protein intolerance becomes more prevalent.

Several clinical reactions to food proteins have been reported in children and adults. Only a few of these have a clear allergic immunoglobulin E (IgE)-mediated pathogenesis. For this reason, the term "food protein intolerance" is usually preferred to "food protein allergy," in order to include all offending specific reactions to food proteins, no matter the pathogenesis.[1] In children, GI symptoms are generally most common, with a frequency ranging from 50-80%, followed by cutaneous symptoms (20-40%), and respiratory symptoms (4-25%).


The major food allergens are water-soluble glycoproteins (molecular weight [MW], 10,000-60,000) that are resistant to heat, acid, and enzymes. Many food allergens have been identified, but milk, eggs, peanuts, tree nuts, fish, soy, wheat, and crustacean shellfishes account for 90% of significant reactions.[2] All these foods contain proteins with a small molecular weight, an abundance of epitopes, water solubility, glycosylation residues, and relative resistance to heat and digestion.[3]

Cow's milk contains more than 20 protein fractions. In the curd, 4 caseins (ie, S1, S2, S3, S4) can be identified that account for about 80% of the milk proteins. The remaining 20% of the proteins, essentially globular proteins (eg, lactalbumin, lactoglobulin, bovine serum albumin), are contained in the whey. Casein is often considered poorly immunogenic because of its flexible, noncompact structure. Historically, lactoglobulin has been accepted as the major allergen in cow's milk protein intolerance. However, polysensitization to several proteins is observed in about 75% of patients with allergy to cow's milk protein.

The proteins most frequently and most intensively recognized by specific IgE are the lactoglobulin and the casein fraction. However, all milk proteins appear to be potential allergens, even those that are present in milk in trace amounts (eg, serum bovine albumin, immunoglobulins, lactoferrin). In each allergen, numerous epitopes can be recognized by specific IgE presence. Cow's milk proteins introduced with maternal diet can be transferred to the human milk. Many studies have focused on the presence of bovine lactoglobulin throughout human lactation. The GI tract is permeable to intact antigens. The antigen uptake is an endocytotic process that involves intracellular lysosomes.

Cow's milk proteins introduced with maternal diet can be transferred to the human milk. Many studies have focused on the presence of bovine lactoglobulin throughout human lactation.

Studies have demonstrated that food allergens are transported in large quantities across the epithelium by binding to cell surface IgE/CD23, which opens a gate for intact dietary allergens to transcytose across the epithelial cells that protect the antigenic protein from lysosomal degradation in enterocytes.

Some antigens can move through intercellular gaps; however, the penetration of antigens through the mucosal barrier is not usually associated with clinical symptoms. Under normal circumstances, food antigen exposure via the GI tract results in a local immunoglobulin A (IgA) response and in an activation of suppressor CD8+ lymphocytes that reside in the gut-associated lymphoid tissue (oral tolerance). Antigen uptake has been found to be increased in children with gastroenteritis and with cow's milk allergy.

Numerous studies have implicated the integrity of the skin and mucosal barrier in protecting against sensitization. In some children who are genetically susceptible, or for other as-of-yet-unknown reasons, oral tolerance does not develop, and different immunologic and inflammatory mechanisms can be elicited.[4] Whether nonimmunologic mechanisms can have a role in the development of specific intolerances to food proteins is still disputed.

Some evidence suggests that reduced microbial exposure during infancy and early childhood result in a slower postnatal maturation of the immune system through a reduction of the number of T regulatory (Treg) cells and a possible delay in the progression to an optimal balance between TH1 and TH2 immunity, which is crucial to the clinical expression of allergy and asthma (hygiene hypothesis). Genetic variations in receptors for bacterial products are likely to be related to allergic sensitizations. On the other hand, intestinal infections may increase paracellular permeability, allowing the absorption of food proteins without epithelial processing. As a consequence, infectious exposures can be an important contributory factor in the pathogenesis of food protein allergies.

Food protein intolerance can be IgE-mediated or non-IgE-mediated. Local production and systemic distribution of specific reaginic IgE plays a significant role in IgE-mediated reactions to food proteins.

Morphologic studies have demonstrated the role of GI T lymphocytes (ie, intraepithelial lymphocytes) in the pathogenesis of GI food allergy. The pathogenetic role of the eosinophils in food-induced eosinophilic GI diseases has not been defined. Vast evidence describes the occurrence of immunoglobulin G (IgG) food protein antibodies. However, their actual role in the pathogenesis of clinically relevant symptoms is, at best, doubtful.

A potentially important factor in the response of the immune system to a specific food antigen is microbiota. In humans, differences in the intestinal flora of allergic versus nonallergic children have been observed.[5]

Non–immune-mediated Reactions

Non–immune-mediated reactions include the following:

  • Disorders of digestive-absorptive process

    • Glucose-galactose malabsorption

    • Lactase deficiency

    • Sucrase-isomaltase deficiency

    • Enterokinase deficiency

  • Pharmacological reactions

    • Tyramine in aged cheeses

    • Histamine (eg, in strawberries, caffeine)

  • Idiosyncratic reactions

    • Food additives

    • Food colorants

  • Inborn errors of metabolism

    • Phenylketonuria

    • Hereditary fructose intolerance

    • Tyrosinemia

    • Galactosemia

    • Lysinuric protein intolerance

Immune-mediated (Food Allergy) Reactions

Immune-mediated (food allergy) reactions include the following:

  • IgE-mediated (positive radioallergosorbent test or skin prick test results)

    • Oral allergy syndrome

    • Immediate GI hypersensitivity

  • Occasionally IgE-mediated

    • Eosinophilic esophagitis

    • Eosinophilic gastritis

    • Eosinophilic gastroenteritis

  • Non-IgE-mediated - Food protein–induced entities (eg, enterocolitis, enteropathy, proctocolitis, chronic constipation)

Autoimmune Reactions

Innate and adaptive immunity reactions (ie, Celiac disease) are noted.



United States

In a national survey of pediatric allergists, the prevalence rate of cow's milk allergy in 1997-1999 was reported to be 3.4%, whereas the prevalence rate of soy protein allergy was 1.1%. During the 10-year period of 1997-2006, food allergy rates significantly increased among both preschool-aged and older children. This trend continued in the following years. According to the data from the National Center for Health Statistics, the prevalence of food allergies increased to 5,1% in 2009-2011 and increased with the increase of income level.[6]


Incidence of food allergy in children has been variously estimated at 0.3-8%, and the incidence decreases with age. Food allergies affect 6-8% of infants younger than 2 years. In a cohort of 1,749 newborns from the municipality of Odense in Denmark who were prospectively monitored for the development of cow's milk protein intolerance during the first year of life, a 1-year incidence of 2.2% was reported.[7]

Varying incidences of specific intolerances have been reported in different countries. Whether these differences are due to genetic or cultural factors is unclear.[8]

To evaluate the prevalence of food allergy among different countries in Europe, the EuroPrevall project was launched in June 2005. Subsequently, the EuroPrevall-INCO project has been developed to evaluate the prevalence of food allergies in China, India, and Russia.[9]


Most of the cases of food protein intolerance can be resolved with dietary management. A few cases of severe anaphylactic reactions to food proteins have been reported. A report from the United Kingdom suggests an incidence of 0.22 severe cases per 100.000 children per year (15% of cases were fatal or near fatal).[10]


No race predilection has been observed.


No sex predilection is known, but males are slightly more frequently affected with eosinophilic gastroenteritis.


Gastrointestinal food protein intolerance is mainly a problem in infancy and early childhood. Cow's milk allergy or intolerance usually develops in early infancy. In most of the cases, the onset of symptoms is closely related to the time of introduction of formula based on cow's milk.

In a prospective study from Norway, the prevalence of atopic dermatitis in the first 2 years was 18.6% with no significant difference between preterm and term children. Adverse reactions to food were found in 15.8% (a similar prevalence in premature and term children). Mode of delivery did not affect prevalence of atopic dermatitis.[11] An example is shown in the image below.

Typical atopic dermatitis on the face of an infant Typical atopic dermatitis on the face of an infant.

An unselected prospective study indicated that 42% of infants who developed cow's milk protein intolerance were symptomatic within 7 days (70% within 4 wk) following the introduction of cow's milk.[12] Cow's milk protein intolerance has been diagnosed in 1.9-2.8% of general populations of infants aged 2 years or younger in different countries of northern Europe, but incidence fell to approximately 0.3% in children older than 3 years.

Protein intolerance is generally believed to remit by age 5 years, when the infant's mucosal immune system matures and the child becomes immunologically tolerant of milk proteins; in most affected children, symptoms resolve by age 1-2 years. However, cow's milk protein intolerance may persist or may initially manifest in older children, demonstrating characteristic endoscopic and histopathologic features; it occasionally recurs in adults.

Studies have suggested increased persistence of food allergies (albeit ones possibly affected by selection bias); possible explanations have been primarily focused on peanut intolerance.[13]




Numerous symptoms can be a consequence of food protein intolerance. GI manifestations are the most common clinical presentation, usually without involvement of other organ systems. Most cases of food protein intolerance in the pediatric population occur in the first months of life as a consequence of cow's milk protein intolerance.

The typical history is that of an infant younger than 6 months who is fed for a few weeks with formula and who then develops diarrhea and, eventually, vomiting. In the case of the common enterocolitis syndrome, the infant can become dehydrated and lose weight. In the rare instance of cow's milk enteropathy, a malabsorption syndrome develops, with growth failure and hypoalbuminemia. On the other hand, the common food-induced proctocolitis syndrome is characterized by diarrhea in a healthy infant without any weight loss.

Food allergic reactions may be divided into quick-onset reactions, which occur within an hour of food ingestion and are usually immunoglobulin E (IgE)-mediated (eg, skin rashes, urticaria, angioedema, wheezing, anaphylaxis), and slow-onset reactions, which take hours or days to develop and are usually non–IgE-mediated.

The most common and specific symptoms of food protein intolerance are as follows:

GI symptoms

  • Oral allergy syndrome: Oral allergy syndrome is a form of IgE-mediated contact allergy that is almost exclusively confined to the oropharynx and is most commonly associated with the ingestion of various fresh fruits and vegetables. Oral allergy syndrome mainly affects adults who have pollen allergy (especially to ragweed, birch, and mugwort) and is caused by cross-reactivity of pollen IgE antibodies with proteins in some fresh fruits and vegetables. Symptoms include itching; burning; and angioedema of the lips, tongue, palate, and throat. The clinical picture is usually short-lived, but symptoms may be more prominent after the ragweed season.

  • Immediate GI hypersensitivity: GI anaphylaxis is defined as an IgE-mediated GI reaction that often accompanies allergic manifestations in other organs, such as the skin or lungs. Bioptic samples show a significant decrease in stainable mast cells and tissue histamine after the challenge. The reaction usually occurs within minutes to 2 hours of food ingestion. Within 1-2 hours, the patient develops nausea, abdominal pain, and vomiting. After 2 hours, diarrhea ensues. Subclinical reactions have been described in children with atopic eczema and food allergy. Poor appetite, poor weight gain, and intermittent abdominal pain are frequent symptoms.

  • Eosinophilic esophagitis

    • Esophageal eosinophilia that persists despite traditional antireflux therapy may represent a sign of allergic esophagitis.

    • Eosinophilic esophagitis was described in early 1990s in adults suffering from dysphagia and in children complaining of severe reflux symptoms refractory to therapy, both associated with an eosinophil-predominant infiltration.

    • Eosinophilic esophagitis occurs in children and adults but rarely occurs in infants and is characterized by chronic esophagitis, with or without reflux. Affected children present with a wide range of symptoms, which are largely age dependent.[14]

    • Children younger than 2 years often present with food refusal, irritability, vomiting, and abdominal pain.

    • Older children, adolescents, and adults present with intermittent vomiting, heartburn, dysphagia for solids, or spontaneous food impaction and failure to respond to conventional reflux medications.

    • In older children, dysphagia, anorexia, and early satiety can help distinguish eosinophilic gastroenteritis from gastroesophageal reflux and correlates with the severity of histologic and endoscopic findings.[15]

    • Occasionally, esophageal strictures develop, apparently due to an esophageal dysmotility.[16]

    • Eosinophilic esophagitis is a chronic disease, with less than 10% of the population developing tolerance to food allergies.[17]


    • Numerous studies have suggested that eosinophilic esophagitis has a strong genetic inheritability. 

  • Eosinophilic gastritis

    • Eosinophilic gastritis that is responsive to elimination diets has occasionally been reported.
    • In children it is usually localized in the antrum and/or fundus, [18] with eosinophilia elsewhere in the GI tract. [19]
    • Symptoms and signs vary, but they are the usual ones for gastritis of different etiologies, such as postprandial vomiting, abdominal pain, anorexia, early satiety, and failure to thrive.
    • Approximately half of these patients have atopic features. The disease is highly responsive to dietary restriction therapies in children.
  • Eosinophilic gastroenteritis: Eosinophilic gastroenteritis is an ill-defined disease that is pathologically characterized by the infiltration of eosinophils in the mucosa of the GI tract. The syndrome has been reported in children of all ages. Diagnosis requires symptoms related to the GI tract and a bioptic sample showing an eosinophilic infiltration. Unfortunately, no clear-cut line can be drawn to distinguish eosinophilic gastroenteritis from other GI diseases and from nonpathologic eosinophilic infiltration of the lower intestine. Eosinophilic gastrointestinal disease seems to precede inflammatory bowel disease in some subsets of children.[20]

  • Food protein–induced enterocolitis syndrome (FPIES)

    • Food protein–induced enterocolitis syndrome describes a complex symptom of profuse vomiting and diarrhea diagnosed in infancy, involving both the small and the large intestine.

    • Food-induced enterocolitis syndrome occurs most frequently in the first months of life. Most cases are observed in infants younger than 3 months.

    • Cow's milk and soy protein are most often responsible.

    • Symptoms include protracted vomiting and diarrhea. The typical presentation is acute. Vomiting generally occurs 1-3 hours after feeding, and diarrhea occurs 5-8 hours after feeding. However, some children have a chronic course, characterized by protracted diarrhea, failure to thrive, and intermittent vomiting.[21]

    • Specific descriptions of the histologic findings are not available because the diagnosis can be made clinically. Some small bowel specimens show mild villous injury with inflammatory infiltration, whereas colonic specimens reveal crypt abscesses and a diffuse inflammatory infiltrate.

    • A similar enterocolitis syndrome has been reported in older infants and children as a consequence of intolerance to different food proteins (eg, eggs, fish, nuts, peanuts, other proteins). Rice can induce severe cases of enterocolitis.[22]

    • Food-specific IgE test findings are typically negative[23] ; atopy patch testing is under investigation. The oral food challenge remains the diagnostic standard in this disorder.[24] Gastric juice analysis can help with diagnosis.[25]

    • During a prospective long-term follow-up study, most patients with infantile food protein–induced enterocolitis syndrome lost intolerance to cow’s milk at age 14-16 months (tolerance rate, 72.7%).[26]  In a cohort of 160 children, the median age for tolerance was 4.7 years for rice, 5.1 years for milk in the absence of milk-specific IgE, and 6.7 years for soy; none of the subjects with detectable milk specific IgE became tolerant to milk during the study.[27]

  • Food protein-induced enteropathy: Cow's milk proteins and soy proteins can cause an uncommon syndrome of chronic diarrhea, weight loss, and failure to thrive, similar to that appearing in celiac disease. Vomiting is present in up to two thirds of patients. Small bowel biopsy findings reveal an enteropathy of variable degrees with villous hypotrophy. Total mucosal atrophy, histologically indistinguishable from celiac disease, is a frequent finding. Intestinal protein and blood losses can aggravate the hypoalbuminemia and anemia that are frequently observed in this syndrome. The nonceliac food-induced enteropathy has been less frequent and less severe in the last 25 years. More recent cases described patients who presented with patchy intestinal lesions. Usually, the syndrome affects infants in the first months of life.

  • Gluten-sensitive enteropathy: See Celiac Disease.

  • Protein-losing enteropathy: Protein-losing enteropathy is a common finding in children with cow's milk protein intolerance. Some infants can present with pronounced protein-losing symptoms after introduction of cow's milk. It has been suggested that mast cell infiltration is related to increased intestinal permeability and protein loss.[28]

  • Food-induced proctocolitis: Food-induced proctocolitis usually occurs in the first few months of life; it is usually considered a delayed non IgE-mediated allergic reaction to food proteins, even if some studies have proposed different pathogenetic mechanisms.[29, 30, 31]  Cow's milk and soy proteins are most often responsible, but 60% of reported infants were exclusively breastfed. In most of the latter cases, a strict maternal diet (including the elimination of all cow's milk–based products from their diets) can resolve the problem. Other allergens, mainly eggs, can be the causative agent in nonresponding infants.[32] Symptoms include diarrhea and blood in the stools. Affected infants generally appear healthy and have normal weight gain. The onset of bleeding is gradual and initially erratic over several days. It then progresses to streaks of blood in most stools, which can elicit suspicion of an internal anal tear. Bowel lesions are generally confined to the distal large bowel. This entity, even if untreated, usually resolves in 6 months to 2 years. Endoscopy is not recommended, unless severe symptoms are present. The duration of the allergic proctocolitis does not seem to induce worsening of the infant's nutritional status.[33]  

  • Chronic constipation due to cow's milk intolerance: Chronic constipation as the sole symptom of intolerance to cow's milk was described in 1993. However, chronic constipation was not considered a feature of cow's milk intolerance until 1998, when an Italian study hypothesized that intolerance to cow's milk can cause severe perianal lesions with pain upon defecation and subsequent constipation in young children.[34] An allergic colitis, with resolution of the symptoms after removal of milk from the diet, was subsequently demonstrated in 4 newborns with constipation. Therefore, in a small subgroup of children with constipation, cow's milk protein intolerance can be the cause of symptoms.

  • Infantile colic

    • Infantile colic is the usual name given to a prolonged pattern of crying or fussing in infants, even if the pathophysiology of this distressing behavior has not yet been elucidated. Numerous theories on the pathogenesis have been published, and many, often conflicting, therapeutic approaches have been suggested.

    • Cow's milk intolerance has been implicated as a cause of colic, at least in some formula-fed infants. Some studies have suggested that an elimination diet that substitutes cow's milk formula with a soy-based formula or a protein-hydrolysate can relieve the symptoms of infantile colic in a significant percentage of cases. In these infants, challenge with cow's milk proteins usually causes a recrudescence of the crying crises. The infants who respond to the elimination diet are usually those with more prolonged crying crises, and they often have a familial history of allergy. Most often, other signs of cow's milk protein intolerance develop in the following weeks or months.

    • Studies including a selected population of infants report percentages of responses to the elimination diet to be as high as 89%. One blind study showed that 18% of infants with colic improved with soy formula, whereas 0% improved in another blind study. Moreover, in most of the responsive infants, the duration of the effect is not sustained, despite an ongoing elimination diet. In any case, true food protein intolerance can only be demonstrated in a small subgroup of infants with colic.

  • Allergic dysmotility: In older children, milk protein intolerance can induce chronic abdominal pain, with an endoscopic finding of lymphonodular hyperplasia.[35]

  • Multiple food protein intolerance of infancy: Some infants are intolerant to cow's milk proteins, soy, extensively hydrolyzed formulas, and a wide range of other food proteins. Most of these children develop symptoms while they are receiving only breast milk. Symptoms remit after feeding with an elemental amino acid–based complete infant formula.

Dermatologic symptoms

  • Symptoms include urticaria, angioedema, rashes, and atopic eczema.

  • Atopic dermatitis is one of the most common symptoms of protein intolerance. Approximately one third of children with atopic dermatitis have a diagnosis of cow's milk protein allergy and cow's milk protein intolerance, according to elimination diet and challenge tests, and about 20-40% of children younger than 1 year with protein intolerance have atopic dermatitis. Most children with atopic dermatitis and protein intolerance develop a complete tolerance in a few years.

  • Umbilical and periumbilical erythema has been related to cow’s milk protein intolerance in a group of 384 Italian infants; this bizarre sign was observed in 36 cases (9.4%), disappeared within the second week on elimination diet, and reappeared within 24 hours after challenge.[36]

Respiratory symptoms

These symptoms include rhinitis and asthma.

General symptoms

Anaphylaxis due to cow's milk protein intolerance is a rare but well-described event. The child, usually a young infant, suddenly becomes pale and cold and sweats. The child usually presents with urticaria or angioedema and goes into shock within minutes after milk ingestion. Anaphylaxis following ingestion of soy protein is exceptionally rare, even though a survey in Sweden identified 4 cases of death caused by soy protein anaphylaxis.[37]

Nonspecific symptoms

Many more nonspecific GI reactions have been ascribed to food allergy, including oral aphthae, pyloric stenosis, and bowel edema and obstruction. For most of these manifestations, a clear correlation with an immune reaction to foods has never been established.


Usually, the GI manifestations are isolated, without any sign of atopic dermatitis, urticaria, rhinitis, conjunctivitis, or wheezing.

  • In proctocolitis syndrome, the child (usually a young infant) appears healthy, without any weight loss or other physical problems.

  • In food-induced GI anaphylaxis, these symptoms (eg, atopic dermatitis, urticaria, rhinitis, conjunctivitis, wheezing) can occur, and, therefore, the child must be checked for the presence of systemic signs of allergic reaction.

  • The infant with enterocolitis syndrome can be dehydrated as a consequence of diarrhea, vomiting, or both. Signs of dehydration include blunted eyes, dry mucous membranes, and hypoelastic skin.

  • In the unusual instances of nonceliac food-induced enteropathy, infants present with signs and symptoms of malabsorption syndrome. Dystrophy, growth failure, edema (hypoalbuminemia), rickets (vitamin D malabsorption), and hemorrhages (vitamin K malabsorption) can all be present.


Many food proteins can act as an antigen in humans. Cow's milk proteins are most frequently implicated as a cause of food intolerance during infancy. Soybean protein is the second most frequent antigen in the first months of life, particularly in infants with primary cow's milk intolerance.



Diagnostic Considerations

Diagnostic considerations include the following:

  • Enteropathy
  • Celiac disease
  • Autoimmune enteropathy
  • Intractable diarrhea of infancy
  • Intestinal infections
  • Enterocolitis syndrome
  • Intestinal infections
  • Inflammatory bowel diseases
  • Proctocolitis
  • Anal fissures
  • Meckel diverticulum
  • Intestinal duplication
  • Intestinal hemangiomas
  • Intestinal infections
  • Inflammatory bowel disease

Differential Diagnoses



Laboratory Studies

Skin test responses to cow's milk or other food proteins and detection of food-specific immunoglobulin E (IgE) antibodies are usually positive in children with IgE-mediated food allergy. However, most of the food protein intolerances are not IgE-mediated. A double-blind, placebo-controlled, oral food challenge is the ideal method for confirming histories of adverse reactions to food proteins. However, this approach is rarely used in clinical practice.

In addition, even double-blind, placebo-controlled challenges can have pitfalls. Encapsulated food extracts minimize the potential to elicit oral, esophageal, or airway reactions and could also result in increased risk because absorption of food might be delayed.[38]

The following tests are indicated in assessing food protein intolerance:

  • Skin testing with food extracts

    • This is often used to screen patients with suspected IgE-mediated food allergies.

    • Skin testing is usually completed by the prick technique, which is a reliable means of detecting IgE antibodies to food in children.

    • A positive skin test result merely implies the presence of food-specific IgE antibodies. A negative skin test result has a high predictive accuracy (estimated to be >95%). Unfortunately, a positive prick test result is a poor predictor of clinical symptoms during food challenges. The positive predictive accuracy widely varies but has been estimated to be lower than 50%. In children younger than 2 years, the negative predictive accuracy of the skin test result is not as good as in older children; however, a positive test result is more likely to be significant.

    • Composition and quality of the extracts are also a significant consideration. Sometimes, testing a drop of the milk that is used to feed the child (eg, formula milk, soy milk, others) is preferable.

  • Serum immunoassays: Serum immunoassays to determine food-specific IgE antibodies are often used to screen for antigen-specific IgE in the patient's serum. Enzyme-linked immunosorbent assays (ELISAs) have been replacing methods that use radiation (eg, radioallergosorbent test [RAST]). Unfortunately, determination of specific IgE involves the same problems as the skin test. A negative test result has a high predictive accuracy with a low sensitivity, whereas a positive test result has a low predictive value.

  • Fecal leukocyte testing: Fecal eosinophils are a significant clue to the diagnosis of allergic colitis. However, although eosinophils predominate after a single acute challenge, the findings on fecal smears at diagnosis are more likely of the inflammatory type, with a predominant population of lymphocytes and neutrophils and a small number of eosinophils.

  • Atopy patch testing: Several studies are evaluating the use of the atopy patch test for delayed intolerances to food proteins. However, no standardized reagents or methods of interpretations are currently available, and the additional diagnostic information in some studies appears marginal.

  • Elimination diets

    • The simplest type of elimination diet is elimination of suspected food antigens from the diet for 2-4 weeks or longer.

    • The duration of diagnostic elimination diet can be reduced to 3-5 days in children with immediate clinical reactions.[38]

    • When no specific allergens can be incriminated, the common food allergens are eliminated from the diet.

    • In breastfed infants, mothers should be encouraged to continue breastfeeding while avoiding all milk and milk products (or other food allergens) in their own diet.[21]

    • In severe and unresponsive cases, consider the use of an elemental diet.

  • Food challenge testing

    • An elimination diet for 10-14 days should precede a food challenge test.

    • Perform a food challenge under medical supervision. A history of food-related anaphylaxis is a relative contraindication to food challenge.

    • The starting dose during an oral milk challenge should be lower than a dose that can induce a reaction and should then be increased stepwise to 100 mL.[38]

    • An open-food challenge (both the physician and the patient are aware of the food) can be accepted when the resulting symptoms can be objectively observed. Use a double-blind, placebo-controlled food challenge (both the physician and the patient are unaware of the food) if the resulting symptoms are only subjective.

    • The oral food challenge has been established as the criterion standard for the diagnosis of adverse food reactions caused by any mechanism. The test is diagnostic of food intolerance when symptoms subside following dietary elimination of the offending food or when symptoms recur within 48 hours after milk provocation. Reactions must be reproducible with repeated elimination and challenge tests. Delayed-onset reactions, a common cause of GI reactions, cause symptoms hours or days after the ingestion, with a timing pattern. In cow's milk protein intolerance, the same original clinical picture tends to recur after the challenge.

    • Very delayed clinical reactions to food challenge, including constipation, wheezing, dermatitis plus constipation, and dermatitis alone, have been reported in a small subgroup of patients. The mean time between the challenge and the onset of a clinical symptom was 13.3 days (range, 4-26 d).

    • Position statements issued by the American Academy of Allergy, Asthma, and Immunology[39] ; National Institute of Allergy and Infectious Diseases (NIAID)[2] ; and European Society for Paediatric Gastroenterology, Hepatology, and Nutrition[40] considered cytotoxic testing, provocative subcutaneous testing, and provocative sublingual tests to be of unproven value.

    • Determination of IgG antibodies or IgG subclass antibodies against CMP has no role in diagnosing CMPA, and therefore, is not recommended.[41]

Specific clinical pictures are provided by the following:

  • Allergic eosinophilic esophagitis

    • Differential diagnosis from gastroesophageal reflux disease (GERD) based on clinical symptoms remains almost impossible in pediatrics.
    • Total serum IgE is within the reference range or slightly elevated.

    • Peripheral eosinophilia is uncommon.

    • Skin test and RAST results are frequently negative.

    • Results of pH testing can be pathologic or normal.

    • Esophageal biopsy reveals infiltration of the mucosa and submucosa with eosinophils.

    • A scoring system of clinical and endoscopic features may help to differentiate eosinophilic esophagitis from GERD (eg, male gender, dysphagia, history of food impaction, absence of pain/heartburn, linear furrowing, white papules). [42]
  • Allergic eosinophilic gastritis

    • The IgE level is elevated in approximately half of patients.

    • Peripheral eosinophilia is present in approximately half of patients.

    • Gastric biopsy findings reveal marked infiltration of the mucosa and submucosa with eosinophils (especially in the gastric antrum).

  • Allergic eosinophilic gastroenteritis

    • Peripheral eosinophilia is commonly found but is not reliable as a diagnostic criterion.

    • Diagnosis requires a biopsy specimen showing an eosinophilic infiltration of the mucosa and submucosa.

    • Unfortunately, no standards for making the histologic diagnosis are recognized. In infants and children without GI symptoms, eosinophilic counts in the gastric fundus and antrum are consistently low, but in the terminal ileum, cecum, and proximal colon, an eosinophil count as high as 30 per high-power field can be detected.

  • Immediate GI hypersensitivity

    • Prick tests and RAST results are usually positive.

    • Fluoroscopy studies show gastric hypotonia, prominent pylorospasm, and subsequent increased or decreased peristaltic activity of the intestine.

  • Enterocolitis syndrome

    • Prick tests and RAST results are usually negative.

    • Stools generally contain occult blood, neutrophils, and eosinophils. The jejunal biopsy findings reveal villous atrophy and infiltration by lymphocytes, eosinophils, and mast cells.

  • Nonceliac food-induced enteropathy

    • The results of both IgE-specific tests (ie, prick, RAST) are usually normal.

    • The findings of the jejunal biopsy are similar to those in celiac disease but usually are less pronounced. A varying degree of villous atrophy is present, with crypt hyperplasia and lymphocytic infiltration of the lamina propria. Often, the lesions have a patchy distribution, observed especially in the last few years. Several studies have detected increased numbers of IgE plasmocytes in biopsy specimens of patients with cow's milk protein intolerance.

  • Food-induced proctitis

    • Fecal leukocytes are usually positive for neutrophils.

    • Eosinophils can be absent.

    • For colonoscopic findings, see Procedures.


The following procedures may be indicated:

Upper and lower GI endoscopies

Usually, the clinical picture and the history are clear, and the recommended diagnostic process (see Laboratory Studies above) does not require performance of endoscopic procedures. However, endoscopy may be part of the differential diagnostic workup in cases in which clinical doubt is acknowledged.

  • Upper endoscopy

    • In cases with eosinophilic esophagitis, various degrees of hyperemia are macroscopically observed. In addition, furrowing of the mucosa, rings, and plaques have been described. Microscopically, eosinophils are observed infiltrating the esophageal wall. Although no pathognomonic histologic characteristics are associated with eosinophilic esophagitis, an eosinophilic count of more than 20 cells per high-power field is considered diagnostic of this recently described entity. Patchy lymphonodular hyperplasia of the GI tract, at any site from the duodenum to the colon, has been suggested to be related to food allergy.

    • In patients presenting with signs or symptoms of enteropathy, an upper endoscopy with duodenal biopsies may be indicated in the diagnostic workup. The microscopic picture of enteropathy can totally overlap with that of celiac disease (ie, partial to total villous atrophy, crypt hyperplasia), possibly with a more common occurrence of a patchy lesion rather than continuous lesion.

  • Lower GI endoscopy (colonoscopy): This procedure is completed in the presence of signs or symptoms of colitis or proctitis (mainly lower GI bleeding) and whenever the diagnosis of food-induced colitis is not obvious. In food-induced colitis, the procedure macroscopically reveals linear erosions and mucosal edema. Bowel lesions are generally confined to the distal large bowel. Microscopically, focal erythema and frequent nodularity with superficial erosions are observed. Eosinophilic infiltration, most prominent in the lamina propria, can be observed in the biopsy specimens but is by no means a constant finding.


In allergic eosinophilic esophagitis, esophageal biopsy reveals infiltration of the mucosa and submucosa with eosinophils.

In allergic eosinophilic gastritis, gastric biopsy reveals marked infiltration of the mucosa and submucosa with eosinophils (especially in the gastric antrum).

In allergic eosinophilic gastroenteritis, diagnosis requires a biopsy specimen that reveals an eosinophilic infiltration of the mucosa and submucosa. Unfortunately, no standards for making the histologic diagnosis are available. In infants and children without GI symptoms, eosinophilic counts in the gastric fundus and antrum are consistently low, but in the terminal ileum, cecum, and proximal colon, eosinophil counts as high as 30 per high-power field can be detected.

In enterocolitis syndrome, the jejunal biopsy reveals villous atrophy and infiltration by lymphocytes, eosinophils, and mast cells.

In nonceliac food-induced enteropathy, the findings of the jejunal biopsy are similar to those in celiac disease but usually are less pronounced. A varying degree of villous atrophy is present with crypt hyperplasia and lymphocytic infiltration of the lamina propria. Often, the lesions have a patchy distribution, observed especially in the last few years. Several studies have detected increased numbers of IgE plasmocytes in biopsy specimens of patients with cow's milk protein intolerance.

Histologic Findings

See Procedures.



Medical Care

The definitive treatment of food protein intolerance is strict elimination of the offending food from the diet.

Breastfeeding is the first choice in infants without lactose intolerance. The mother should eliminate cow's milk (and eventually eggs and fish or other implicated foods) from her diet.

As many as 50% of children affected by cow's milk protein intolerance develop soy protein intolerance if they are fed with soy-based formulas. Therefore, soy-based formulas should not be used for the treatment of cow's milk protein intolerance. Use complete milk protein hydrolysates in infants who cannot be breastfed. Partially hydrolyzed formulas are absolutely not indicated in children with cow's milk protein intolerance. Occasionally, children may develop intolerance toward complete hydrolysated formulas. In these cases, use amino acid–based formulas, which are now widely available and are balanced in trace elements and vitamins.

Eosinophilic gastroenteritis can show clinical and histologic improvement after oral corticosteroid therapy. Topical steroids, administered as inhaled corticosteroids, have also shown beneficial effect.

Consensus recommendations for eosinophilic esophagitis in children and adults by a multidisciplinary group of experts[43] and management guidelines by ESPGHAN[40] agree that treatment involves dietary therapy of 3 possible regimens: strict use of amino acid-based formula, dietary restriction based on allergy testing, or dietary restriction based on eliminating the most likely food antigens. 

The elemental diet (aminoacid-based formulas) and 6-food elimination diet (withdrawing cow´s milk, wheat, egg, soy, nuts and fish/seafood for 6 weeks) appear to be the more efficacious treatment. However, the high level of restriction (and the need for multiple endoscopies) discourage many patients. A new strategy, starting with a 2-food elimination diet (without animal milk and gluten-containing cereals), followed by a 4-food (animal milk, gluten-containing cereals, egg, legumes), and eventually by a 6-food elimination diet in case of failure, may be better tolerated.[44]

Swallowed topical steroids (fluticasone propionate or oral viscous budesonide for a minimum of 4 wk and a maximum of 12 wk) should be considered as a treatment option either alone or in combination with an elimination diet.[45]  Treatment with cromolyn sodium, leukotriene receptor antagonists, anti-TNF agents, antibodies against immunoglobulin E (IgE), or interleukin (IL)-5 (reslizumab) and immunosuppressive agents cannot be recommended for treatment in children with eosinophilic esophagitis.[46]  

Administration of food allergens as immunotherapy carries a greater risk of adverse and potentially severe allergic reactions compared with the administration of inhalant allergens.[47, 48] Based largely on the clinical experience published in European trials, the general impression is that food allergen exposure through the oral or sublingual routes is less risky than through the subcutaneous route, but this perception has yet to be definitively demonstrated.

Recombinant monoclonal humanized anti-immunoglobulin E (IgE) therapy has been approved for the treatment of asthma with associated environmental allergies, but the response can vary with food allergies.

A 9-herb formula based on traditional Chinese medicine is currently under investigation as a treatment for food allergy.[49]

Although probiotics might have a role in the treatment of food allergy by promoting gut barrier integrity, suppressing intestinal inflammatory responses, and inducing IgA production and tolerogenic immune responses, no evidence suggests that such an approach is effective for induction of tolerance in the clinical setting.[38]  

Current evidence indicates that probiotic treatment does not modify the natural course of food allergy.[42]  However, 4 children treated with Lactobacillus rhamnosus without food restrictions showed a resolution of their symptoms within 7-28 days.[50]    



Medication Summary

Topical or orally and intranasally inhaled corticosteroids are used to treat dermatologic or respiratory symptoms associated with protein intolerance. Antihistamines and inhaled bronchodilatators are used as appropriate for mild cases of immediate hypersensitivity. In severe anaphylactic reactions, intramuscular epinephrine can be life-saving.


Class Summary

These agents have anti-inflammatory properties and cause profound and varied metabolic effects. They modify the body's immune response to diverse stimuli.

Triamcinolone topical (Aristocort)

Treats inflammatory dermatosis responsive to steroids. Decreases inflammation by suppressing migration of polymorphonuclear leukocytes and reversing capillary permeability.

Hydrocortisone topical (Cortaid, Dermacort, Westcort, CortaGel)

Adrenocorticosteroid derivative suitable for application to skin or external mucous membranes. Has mineralocorticoid and glucocorticoid effects resulting in anti-inflammatory activity.

Corticosteroids, Inhalants

Beclomethasone, inhaled (Qvar, Qvar RediHaler)

Inhibits bronchoconstriction mechanisms. Produces direct smooth muscle relaxation. May decrease number and activity of inflammatory cells, in turn decreasing airway hyperresponsiveness and inflammation.




The following should be considered in food protein intolerance:

  • Infants with elevated cord serum immunoglobulin E (IgE) and a positive family history of atopy are at risk for the development of atopic disease.

  • In some infants at high risk, exclusive breastfeeding with delayed introduction of solid foods until the infant is aged 6 months may delay or possibly prevent the onset of food allergy.

  • Because small amounts of food antigens ingested by the mother are excreted in breast milk, avoidance of allergenic foods by lactating mothers can be recommended; however, randomized, controlled trials of the elimination of food allergens from the diet of mothers during pregnancy and breast-feeding or from the diet during the first year of life have not shown reductions in the risk of IgE-mediated food allergies in children at age 7 years.[51] Some studies suggest that infants who are exposed to food allergens early are less likely to have food allergies.[52] Guidelines for nutritional interventions to prevent atopic disease have been established by the American Academy of Pediatrics (AAP).[53]

  • Two systematic reviews in Cochrane Database did not find any evidence to support feeding infants with hydrolyzed formula or soy protein formula for the prevention of allergy or protein intolerance.[54, 55]

  • In the early 2000s a series of international guidelines recommended late, restricted weaning, especially in high-risk infants. A position statement from the American Academy of Pediatrics (AAP) and a document from the American College of Allergy, Asthma, and Immunology recommended withholding cow’s milk until age 1 year, eggs until age 2 years, and peanuts, and tree nuts and fish until age 3 years, particularly in high risk children. [56]  
  • Numerous subsequent prospective studies have failed to demonstrate an association between early introduction of cow's milk or complementary foods and food allergy. Conversely, an increased risk of atopic dermatitis and allergic sensitization has been associated with delayed introduction of complementary foods. Tolerance to food allergens appears to be driven by regular exposure to proteins during a ‘‘critical early window’’ of development. Therefore, concern is mounting that the recommended practices of delaying the introduction of complementary foods beyond age 6 months may increase, rather than decrease, the risk of immune disorders. Current evidence suggests that the timing of this window is most likely to be allocated between age 4 months (around 17 wk) and age 7 months. [57, 58, 59]  
  • Two position papers from ESPGHAN and AAP raise questions about the benefit of delaying the introduction of solid foods that are thought to be highly allergic beyond age 4-6 months. [58, 59] Therefore, avoidance or delayed introduction of allergenic foods for the purpose of avoiding allergies is not recommended.  
  • The intestinal microflora, or microbiome, interacts with the mucosal immune system, and, in germ-free mice, does not develop a normal oral tolerance. The intestinal flora of children with atopy has been found to differ from that of controls. These observations suggest that the normal flora can play a role in the prevention of food allergies, probably due to its substantial effect on mucosal immunity. Probiotics can potentially modulate the immune response, mainly by stimulation of Th1 cytokines that can suppress Th2 responses.  
  • A potential role for probiotics can be hypothesized. Extremely encouraging data are accumulating in this area, even if conflicting results are still reported. Administration of L rhamnosus to pregnant and lactating mothers and their offspring for the first few months of life seems to be safe and was shown to be effective in preventing the development of eczema in 50% of children at high risk for food allergy throughout the first 7 years of life. 


See the list below:

  • Food-induced intolerance is most often a temporary disease. Most children can resume consumption of the offending antigen after 1-4 years of elimination diet.

  • Thirty-nine infants with proven cow’s milk protein intolerance from a cohort of 1,749 newborns from the municipality of Odense in Denmark had a good overall prognosis, with a total recovery of 56% at age 1 year, 77% at age 2 years, 87% at age 3 years, 92% at ages 5 and 10 years, and 97% at age 15 years.[7] In children younger than 10 years, 41% developed asthma, and 31% developed rhinoconjunctivitis.

  • In food-induced proctocolitis, symptoms generally clear within several days. However, complete resolution of occult bleeding may take as long as 6 weeks.

  • Infants with IgE-mediated cow’s milk proteins allergy have a higher risk for development of allergy against environmental inhalant allergens.