Key Points
- •
A food allergy is an adverse health effect arising from an immune response that occurs reproducibly on exposure to a given food, whereas a food intolerance is an adverse effect due to a nonimmunologic response, e.g. metabolic, pharmacologic or toxic.
- •
Food allergies may be ‘immediate’ (IgE-mediated) or ‘delayed’ (non-IgE-mediated) in onset and induce a variety of symptoms involving the skin, respiratory or gastrointestinal tracts and/or cardiovascular system.
- •
A thorough and detailed history is the most important part of the evaluation and determines which laboratory tests should be ordered, which food challenges and treatments may be required, and the education that will be needed regarding the results.
- •
Prick skin tests and food-specific IgE levels confirm sensitization and provide some evidence on the probability of clinical allergy, but alone are never adequate to make the diagnosis of food allergy.
- •
The oral food challenge remains the ‘gold standard’ for diagnosing food allergy.
Introduction
It is now 40 years since the first double-blind placebo-controlled food challenges were performed and demonstrated that histories of adverse food reactions could be objectively confirmed or refuted. Recently, new diagnostic tools have become available that may decrease the number of challenges required for accurate diagnosis and management. In addition, treatments for food allergy are being developed and studied, and while they are not quite ready for clinical use, these developments are exciting and offer hope that one day a treatment will end the impact of food allergy on patients’ quality of life.
The definitions of terms in this discussion are derived from and consistent with the 2010 Guidelines for the Diagnosis and Management of Food Allergy in the United States: summary of the NIAID Sponsored Expert Panel Report, and the recent Practice Parameter Update.
‘ Food allergy is defined as an adverse health effect arising from a specific immune response that occurs reproducibly on exposure to a given food.’ These immune responses may be IgE-mediated, non-IgE-mediated or an apparent mixture of multiple mechanisms. The term food intolerance is used to designate a nonimmune mediated reaction that may include metabolic, pharmacologic or toxic mechanisms. Another important term is sensitization , which indicates that individuals may have demonstrable IgE or other antibodies or antigen-reactive cells in the absence of clinical symptoms. An immune-mediated food allergy requires both the presence of sensitization and clinical responsiveness when the food is ingested.
Prevalence
Current literature suggests that food allergy may affect up to 10% of the population, but probably not more than that percentage. Studies of prevalence have certain limitations, making them difficult to compare. In children, a 2009–2010 study estimated that 8% of the study population of children have food allergy. In a Canadian study, after adjusting for improbable reports of food allergy, it was estimated that 6.7% of the overall population, 7.1% of children and 6.6% of adults, had food allergy. Since these conclusions were based upon self-reporting, it is clear that they are estimates at best. McGowan et al used the National Health and Nutrition Examination Study (NHANES) 2007–2010 data based upon specific food allergen IgE measurements and a specific set of definitions to arrive at an estimated prevalence of food allergy among children of 6.5%.
IgE-Mediated Symptoms ( Box 41-1 )
Cutaneous Symptoms
Food allergy-induced skin symptoms generally fall into two main categories: atopic dermatitis and urticaria. Urticaria typically begins promptly after the ingestion of a known food allergen, is usually diagnosed early and the culprit food determined. Chronic urticaria is rarely due to a food allergen. Rarely, food additives have been reported to cause chronic urticaria in adults, but there are no systematically confirmed reports in children.
IgE-Mediated
Immediate (gastrointestinal, respiratory, cutaneous, ocular, cardiovascular, anaphylactic)
Immediate and late-phase (atopic dermatitis, allergic gastrointestinal disorders)
Oral allergy syndrome or pollen-food allergy syndrome
Non-IgE Immune-Mediated
Celiac disease, dermatitis herpetiformis
Food protein-induced gastrointestinal illnesses
- •
Food protein-induced enterocolitis
- •
Eosinophilic esophagitis, gastroenteritis (allergic)
- •
Allergic colitis/proctocolitis
- •
Food protein-induced enteropathy (milk, soy, others)
- •
Food-induced pulmonary hemosiderosis (Heiner’s syndrome)
Non-Immune-Mediated
Toxic reactions
Toxic reactions (food poisoning, e.g. scombroid fish poisoning)
Non-toxic reactions
Intolerances
Carbohydrate malabsorption (e.g. lactase deficiency, fructose deficiency, sucrase-isomaltase deficiency)
Psychological reactions (strongly held beliefs)
Atopic dermatitis has been shown to be exacerbated by food allergies in numerous, carefully controlled studies using double-blind, placebo-controlled food challenges. Food allergic reactions may trigger an eczematous rash in 30–40% of children with moderate-to-severe atopic dermatitis. In some situations, the onset of symptoms is subtle and somewhat delayed, with irritability and then itching being the first symptoms to appear, followed by erythema and/or urticaria preceding the more typical erythema and morbilliform eruption that may be most prominent the day after the offending food is consumed. There does not appear to be a single pattern of presentation, but careful observations by families can often make the connection, especially when parents are instructed on the typical presentation. The mechanisms of these reactions are considered in detail in Chapter 47 .
Respiratory Symptoms
Respiratory symptoms of a food allergic reaction include those in the upper respiratory tract – sneezing, nasal pruritus, rhinorrhea and congestion, and periocular pruritus and tearing. In the lower respiratory tract, symptoms and signs include stridor, hoarse voice, cough, dyspnea and wheezing. It is important to note that a hacking staccato cough may be a sign of impending laryngeal obstruction without other symptoms and may lead to abrupt airway closure. Asthma is infrequently the sole manifestation of an allergic reaction to food. However when a patient with asthma has symptoms that are not responding in the usual fashion to treatment, a food reaction should be considered and the treatment approach altered to include injected epinephrine. Food allergy in individuals with asthma may predispose them to more severe episodes and may be a risk factor for more severe and fatal asthma.
Gastrointestinal Symptoms
Immediate-onset gastrointestinal symptoms include nausea, abdominal pain and cramps, vomiting and diarrhea. Nausea and vomiting are often immediate – while the food is being consumed, raising the suspicion of food allergy, especially in an individual with a known food allergy. In individuals without known food allergy, these symptoms should raise the suspicion of a food allergy as often as ‘food poisoning’ is suspected. It should be noted that the gastrointestinal symptoms may not be accompanied by skin manifestations. Rapid resolution of gastrointestinal symptoms and return of appetite are frequently noted after a gastrointestinal food reaction. Diarrhea may occur immediately or be delayed for a few hours.
The issue of colic as a gastrointestinal food allergic reaction in infants remains controversial. With newer feeding guidelines suggesting that delaying food introduction may increase food allergy, it is often difficult to determine whether or not to try elimination diets. However, when infants and families are under significant distress, a brief trial of dietary elimination may be warranted.
Pollen/Food Allergy Syndrome (also see Chapter 46 )
Pollen-food allergy syndrome (oral allergy syndrome) is a common disorder in which oral symptoms, itching of the throat and occasionally mild oral edema occur immediately upon the ingestion of certain foods, most commonly raw fruits and vegetables, but also certain nuts, e.g. hazelnuts and peanuts can trigger these symptoms. These complaints are due to specific IgE antibodies directed to aeroallergens that cross-react with certain food proteins. It is often useful to perform skin tests with fresh fruits and vegetables to confirm the diagnosis, and component protein testing for hazelnut (Cor a 9 and 14) and peanut (Ara h 8) will identify IgE to the birch pollen cross-reactive protein, Bet v 1. This constellation of symptoms is often present in children, but it is often unrecognized unless patients with pollen allergy (and/or their parents) are specifically queried about the presence of oral symptoms with certain foods.
Chronic Constipation
A few controlled trials suggest that this is worth considering in youngsters having persistent constipation issues, but this remains controversial.
Eosinophilic Esophagitis (see Chapter 45 and below)
Many children with eosinophilic esophagitis (EoE) have an IgE-mediated food allergy and atopy, but the underlying immunopathogenic mechanism of EoE is not IgE-mediated and routine allergy tests are generally not helpful in identifying foods provoking symptoms.
Cardiovascular Symptoms
Allergic reactions to foods that involve the cardiovascular system in children usually appear as respiratory compromise first and then progress to a drop in blood pressure and shock, as contrasted with adults who may have the sudden onset of cardiovascular symptoms before any other symptoms occur.
Anaphylaxis (see also Chapter 58 )
The working definition of anaphylaxis is ‘a serious allergic reaction that is rapid in onset and may cause death.’ The clinical criteria for diagnosing anaphylaxis are outlined in Chapter 58 . Acceptance and use of these criteria should aid emergency responders in the rapid identification of anaphylaxis and prompt the institution of resuscitative measures, as validated in recent studies.
Non-IgE Immune-Mediated Reactions to Food
Celiac disease is an autoimmune process that occurs when antibodies directed to gluten cross-react with epithelial cells in the gastrointestinal tract. When gluten-containing foods are removed from the diet, the gastrointestinal lesion resolves. An associated but less common condition is dermatitis herpetiformis that is often mistaken for atopic dermatitis. Virtually all individuals with celiac disease exhibit HLA-DQ2 or DQ8 genetic haplotypes.
The eosinophilic gastrointestinal disorders (see Chapter 45 ), especially EoE, appear to be increasing. In the majority of children, food allergy is a significant trigger; however, it is not IgE-mediated and therefore current diagnostic methods for detection of IgE (skin tests and specific serum antibody measurements) are generally not helpful. A careful and detailed history may be the most important means of raising suspicion of the disease and prompting referral of the youngster for endoscopy. A pattern of seasonal exacerbation in some individuals raises the possibility of swallowed aeroallergens as triggers. It is often difficult to monitor the effectiveness of food elimination diets because there are no noninvasive tests to examine the esophagus for a response.
Another gastrointestinal syndrome that is immunologically mediated is food protein-induced enterocolitis syndrome (FPIES; see also Chapter 44 ). A careful history elicits a pattern of repetitive vomiting that is delayed by about 2 hours after the ingestion of culprit foods. Characteristics of FPIES that distinguish it from IgE-mediated gastrointestinal reactions are the delayed onset of symptoms, the lack of immediate recovery after the vomiting, and the continuous pattern of vomiting, pallor and hypotension in about 15% of cases. There is a helpful website for both parents and providers (fpies.org). Food protein-induced proctocolitis is characterized by gross or occult blood in the stools with an otherwise healthy appearing infant (see Chapter 44 ).
Non-Immunologic Reactions
Toxic Reactions
A number of toxic reactions have been described that could be confused with allergic reactions to food. Food poisoning due to bacterial contamination commonly provokes nausea, abdominal pain and often profuse diarrhea. Scombroid fish poisoning, due to histamine in poorly prepared histidine-containing fish, is less common and can more easily mimic an allergic reaction, including triggering skin changes that are not caused by bacterial food poisoning. These skin changes may include flushing, urticaria and angioedema. Respiratory symptoms may occur due to the large amount of histamine present.
Non-Toxic Reactions
Auriculotemporal syndrome (Frey’s syndrome) is triggered when foods that increase salivation cause a flushing reflex through the auriculotemporal branch of the trigeminal nerve resulting in a ‘strap-like’ rash on both sides of the face. Gustatory rhinitis triggers rhinorrhea due to the ingestion of spicy foods.
Lactose intolerance due to lactase deficiency is the most common carbohydrate malabsorption condition. When there is insufficient lactase in the intestinal mucosa, diarrhea and bloating ensue, and the condition may be confused with milk allergy. Depending upon the degree of lactase deficiency, some patients may tolerate small quantities of milk products without symptoms. Chronic diarrhea in young children may be due to carbohydrate malabsorption caused by fructose in fruit and especially fruit juice. The diarrhea, which often has an acrid smell, may be accompanied by a scalded skin appearance in the perianal and diaper area in the youngest children.
Psychological Reactions
Some parents harbor strongly held beliefs about specific foods that trigger various symptoms in their children, including behavioral changes. These beliefs are usually imposed on children by their parents and may lead to food aversions. Clinicians must be vigilant to ensure these beliefs and dietary restrictions do not lead to malnutrition or deficiency in specific nutrients. Occasionally, Münchausen syndrome by proxy must be considered, most often related to behavioral changes or other subjective symptoms.
Evaluation
History
A thorough and detailed history is the most important part of the evaluation and will determine which specific laboratory tests to order, which food challenges and treatments may be required, and the education that will be needed regarding the results and avoidance of food triggers. The details to be ascertained include a detailed description of the following: symptoms that have been observed including the sequence of those symptoms; timing from onset of symptoms to their resolution; number of events that have occurred for each suspected food; possible ingestion of the food without symptoms; quantity of food eliciting symptoms, including the least amount (threshold) that has triggered symptoms if there has been more than one event; and associated factors such as exercise (in food-dependent exercise-induced anaphylaxis), medication (especially antireflux medication) and alcohol ingestion accompanying the suspected food.
A history of anaphylaxis or a severe reaction increases the need for accurate details that include getting the ingredients of a meal from the facility (restaurant, home or school) where the reaction occurred. This may lead to suspicions about less obvious culprits, especially spices. Emergency department records may be helpful. It is also important in situations where wheezing has been part of the reaction, to inquire if the asthma symptoms responded in the usual manner for that individual. If not then a food might have been the trigger.
Physical Examination
A complete physical examination should be done in children with a history of a food allergic reaction; however, the exam is usually normal unless the reaction is occurring acutely. The major exception is atopic dermatitis, which is a chronic condition that may exacerbate during the acute reaction. Other stigmata to observe include a possible abnormal abdominal examination, signs of malnutrition, or significant failure to thrive in young children placed on a restricted diet.
Laboratory Studies
Skin Testing
Skin testing by the prick/puncture technique is an easy to perform, cost-effective method for identifying sensitization to a food and for determining the probability that a food challenge is likely to be helpful. Food allergens eliciting wheal diameters of at least 3 mm or larger than the negative control are considered positive test results. Negative skin tests have a high negative predictive accuracy, thus usually excluding food allergy to common foods. The negative predictive accuracy for children younger than 3 or 4 years of age tends to be lower than for older children.
Food extracts that elicit a positive result in the absence of a strong history of clinical reactivity typically have a positive predictive accuracy of less than 50% and cannot be considered diagnostic of symptomatic food allergy. Some studies suggest that larger skin tests (≥ 8-mm wheals for some foods) correlate better with symptomatic food allergy, but there is no correlation between skin test size and severity of reactions. If there is a history of a convincing allergic reaction to a food and the skin test is positive, the test may be viewed as diagnostic. There have been rare reports of adverse reactions to intracutaneous skin tests.
Skin test outcomes can be variable depending on a few factors: reagents and devices used for testing, experience of the testing personnel and the interpretation of the test results. A strongly positive history incriminating a specific food in the face of a negative skin test must be evaluated further, e.g. food-specific serum IgE determination and/or physician-supervised oral food challenge.
Skin tests must be selected judiciously based upon the history rather than performing ‘panels of food skin tests’. Selection of numerous foods to be eliminated from the diet based on large numbers of poorly selected skin tests may lead to diets that are difficult for families to follow and may eliminate foods that are clearly tolerated, making adherence to the diet poor. In rare instances these diets can be so strict as to be nutritionally inadequate.
Commercial skin test extracts vary considerably in allergen content. It is often very useful to use fresh fruits and vegetables for skin testing by the technique referred to as ‘prick to prick’. In this technique the fresh food is ‘pricked with the skin test device and then the skin is pricked immediately’. The results of these tests are very helpful when positive, but less so when negative.
Limitations of skin testing include a number of variables: (1) commercially prepared extracts often lack labile proteins responsible for IgE-mediated sensitization to most fruits and vegetables (as noted above); (2) skin testing on skin surfaces that have been treated with topical steroids may induce smaller wheals than those measured on untreated skin; (3) negative prick skin tests with commercial extracts that do not confirm convincing histories of food reactions should be repeated with the fresh food before concluding that IgE is absent; and (4) long-term, high-dose systemic steroid therapy may reduce allergen wheal size.
Intradermal skin testing for food is not recommended because of its high false positive rate and its occasional association with systemic reactions. However, intradermal skin tests have been found to the useful in reactions to beef, pork and lamb due to reactivity to the meat proteins and galactose-α-1, 3-galactose.
In Vitro Testing
Numerous studies have shown that specific serum antibody levels correlate well with the outcome of oral food challenges, especially for peanut, egg, milk and tree nuts, but there is less information on fish, shellfish and a few other foods. As with skin testing, detectable antibody in an immunoassay gives probability information on the likelihood of a reaction to a suspected food, but the history and food challenge remain crucial. Two early studies established ‘cut-off values’ giving 95% decision points, which suggested that values exceeding these levels obviated the need for a food challenge. Since then, numerous other studies have been performed in various populations and with children of various ages. Recent studies suggest that longitudinal monitoring of specific serum antibody levels may be useful in deciding when a child may have outgrown a particular food allergy and when a challenge is likely to be helpful in identifying the resolution of a food allergy. The rate of fall of specific serum antibody levels may be a useful predictor of the resolution of a food allergy. These decision points have also been used as ‘cut-off’ values to determine a reasonable level for doing food challenges based on the likelihood of resolution of a food allergy. Two recent studies have formulated predictive curves for the probability of resolution of egg and milk allergy in young children.
In vitro measurements are preferred in a number of situations: (1) patients with extensive dermatographism; (2) patients with extensive skin disease (atopic dermatitis or urticaria); (3) patients who cannot discontinue antihistamines; and more recently (4) use by nonspecialists who do not perform skin testing to evaluate children for potential food allergy.
There is no consensus on whether skin tests or specific serum antibody levels are most sensitive. However, at least one study has found that prick/puncture skin tests and immunoassays have similar sensitivities and specificities when compared with double-blind, placebo-controlled food challenges. At present, most allergists use the two tests together to decide whether or not to do challenges (see below), and the probability of the challenge being positive for a particular food ( Table 41-1 ).
