Food allergy and atopic dermatitis

Food allergy is defined as an adverse health effect arising from a specific immune response — immunoglobulin E-mediated — that occurs reproducibly on exposure to a given food.

IgE-mediated reactions to food affect up to 6% of children aged younger than 5 years and are more frequently seen in children with atopic dermatitis (AD). In addition, the younger the child and the more severe the AD, then the greater the likelihood of food allergy. The most common foods that cause food allergy are milk, egg, peanut, wheat, soy bean, tree nuts, fish and shellfish. Milk, eggs and peanuts account for most food allergic reactions in young children, and peanuts, tree nuts and seafood account for most in teenagers and adults. This editorial will deal with food allergy in children who have AD. Because of space limitations, I will not discuss mechanisms or non–IgE-mediated food reactions.

Several studies report on the co-occurrence of other allergic conditions in patients with food allergy, with 35% to 71% of food allergic children having AD. In patients with both AD and food allergy, 44% had allergic rhinitis and asthma, and 27% had allergic rhinitis. There is a high prevalence (30%-40%) of food allergy in those with moderate to severe AD.

Children with moderate to severe AD frequently have positive immediate skin tests or serum IgE directed to food allergens.

Up to 37% of children aged younger than 5 years with moderate to severe AD will have food allergen-specific serum IgE (sIgE) to food allergens (milk, egg, wheat, soy bean, peanut and fish), with a positive food challenge or a convincing history in many of them. These studies suggest that food allergy and significant AD will occur often in the same patient, and that early onset of AD is associated with increased sensitization to foods.

Child with presumed food allergy

Evaluation of a child with suspected food allergy begins with the medical history and physical examination.

Utilizing a detailed medical history may be revealing, but history alone cannot be diagnostic of food allergy. A systematic review of the patient’s dietary history should be conducted, with particular attention to associations with the suspected food, including amount ingested, the timing of symptoms, its reproducibility, treatment and outcome. My experience has been that young children who may be allergic to a food will often refuse to eat it. IgE-mediated food reactions in children occur mostly from ingestion but can also occur from cutaneous exposure or inhalation. Symptoms usually occur right after ingestion within minutes, but they can occur up to 2.5 hours, but never days or weeks.

Multiple target organs can be involved, although the skin and gastrointestinal tract are most common. Be aware when one obtains the history that the details of the reaction are subject to recall. And the tricky part is that trace amounts of foods can contaminate other foods, and hidden ingredients may be overlooked. Reading labels and reviewing these labels with the parents and applying them are crucial.

Important questions to ask include:

• What were the symptoms?
• When did they occur with regard to exposure of the food? Has the food ever been eaten without symptoms?
• Have the symptoms occurred without exposure to the food?
• How long did the symptoms last after treatment?

It is important to note that most studies demonstrate that up to 90% of “food allergies” are not actually allergies. Therefore, it is important that all suspected food allergy be confirmed by evaluation, including tests for sensitization and possibly food challenge.

Two systematic reviews found that the prevalence of food allergy based on self-reported symptoms were significantly higher when the diagnosis was based on sensitization alone, sensitization with symptoms, or by double blind, placebo-controlled food challenge.

Physical exam is always important when evaluating a patient but is never diagnostic of food allergy. Basing the diagnosis of food allergy on either history or physical examination alone may lead to the wrong diagnosis.

Methods to identify the culprit

Skin-prick testing

Performing skin-prick testing (SPT) to assist in the identification of the provoking food is helpful, but alone cannot be considered diagnostic of food allergy. It has a low positive predictive value to diagnose food allergy. Nevertheless, SPTs can detect the presence of food-specific IgE antibodies, but many patients with AD have food-allergen-specific serum IgE (sIgE) without clinical food allergy.

Clinicians must remember that in a patient with confirmed food allergy, SPTs are valuable in identifying the food or foods responsible for IgE-mediated food allergy. They provide immediate results and currently are the most commonly performed procedure in the evaluation of IgE-mediated food allergy. Such testing should be done by a board-certified allergist/immunologist. A positive test is considered a wheal with a 3-mm diameter or more than a negative control. One must measure the size of the wheal and erythema and do the test with negative and positive (histamine) controls. The larger the mean wheal diameter, the more likely that the food allergen will be of clinical significance. It is unusual to have a negative SPT in patients with IgE-mediated food allergy. However, if the history is highly suggestive, then one should skin test with the actual food and not the commercially produced food allergens. And if that’s still not positive, then an oral food challenge is indicated.

Intradermal skin testing should not be done to evaluate a child for the diagnosis of food allergy. There is no evidence to support its diagnostic abilities, there is a higher degree of false positivity, and there is always a risk for systemic reaction.

Total serum IgE

Total serum IgE is not at all helpful in evaluation of the food allergic child.

Food allergen-specific serum IgE

SIgE tests are useful for identifying foods that may provoke IgE-mediated food allergic reactions. Their defined cutoff levels for some common food allergens are shown in the Table. It is important to remember that sIgE measures IgE in the serum and SPTs reflect IgE bound to cutaneous mast cells. At times, the results do not correlate. Serum testing, however, is very useful in the child with extensive AD because there may not be adequate normal skin to test. SIgE testing originally was measured using radioallergosorbent test (RAST), but this test has been replaced by the more sensitive fluorescent enzyme-labeled assays. The one most commonly used now is the ImmunoCAP (Phadia).

As with the skin tests, the presence of sIgE reflects allergic sensitization but not necessarily clinical allergy. However, a number of studies comparing the amount of sIgE with oral food challenges have reported that the greater the levels of sIgE, the higher the probability that ingestion of the food will lead to a reaction.

My own personal experience is that combining the size of the wheal on a SPT and the amount of sIgE is very helpful in not just diagnosing food allergy, but also in determining whether a food challenge, either double blind or open, is indicated.

SPTs and sIgE both detect sensitization to foods that may be responsible for IgE-mediated food allergic reactions. However, these tests have poor specificity and relatively poor overall correlation with clinical reactivity. Therefore, when used alone, they do lead to a significant amount of over-diagnosis of clinical allergic reactivity.

Oral food challenges

Oral food challenges remain the gold standard to diagnose food allergy. Testing may be performed either as an open, single blind or double blind, placebo-control food challenges. These challenges should only be performed under strict medical supervision and in a setting properly equipped to treat potential severe reactions. The starting dose is kept low, so as not to trigger severe reactions. Doses are increased over 15 to 20 minutes, until a defined dose is reached.

Literature review of food allergy, AD

Eigenmann and colleagues conducted a prospective study to determine the incidence of IgE-mediated food allergy in patients referred for evaluation of AD (63 children, median age of 2.8 years).

These patients had sIgE antibodies to milk, egg, wheat, soy, peanut and fish, determined by College of American Pathologists (CAP) immunoassay (41 patients had positive values, and 22 patients had negative values). Of the 41 patients with positive IgE values, 19 had food challenges and 11 were positive. Overall, 23 of 63 patients (37%) exhibited clinically significant IgE-mediated clinical reactivity to food. The authors concluded that in patients with moderate-severe AD, evaluation for food allergy should be performed.

In another study, 74 Swiss children were evaluated (median age of 2.5 years) with AD for the prevalence of food allergy. Initially, patients were skin-prick tested to common food allergens and to suspected foods based on clinical history. Thirty patients had negative SPT; 19 patients were diagnosed with food-induced anaphylaxis based on clinical history and/or CAP values. Altogether, food allergy was noted in 33.8% of the study group, with egg, milk and peanut being the most common causative foods.

In another study, 487 infants were followed from age 6 months to 1 year. A total of 141 infants (29%) were classified as having AD. Ninety patients had IgE-mediated food allergy, with 56% having AD. The prevalence of IgE-mediated food allergy increased as the severity of AD increased. For the group with severe AD, 69% had IgE-mediated food allergy (by SPT). No food challenges were done.

In a multicenter, international study, 2,184 participants (mean age of 17.6 months) with AD were studied, and 68% of these patients were noted to have moderate-severe eczema.

The earlier the age of onset and the greater the severity of AD, the more likely it was for patients to have high levels of IgE-mediated food sensitivity. Sensitivity to milk, egg and/or peanut was greatest in patients whose AD developed in the first 3 months of age.

Garcia and colleagues studied 44 infants (mean age of 7.5 months) with AD, and 27% had positive food challenges. Food sensitization was noted in 61% of the study population. Food sensitization was noted in 43% of those with mild AD, 68% of those with moderate AD and 100% of those with severe AD. Egg white was the most commonly observed food (61%).

Multiple studies have shown that at least 30% of patients with severe AD have food sensitization. Addressing food allergies and initiating elimination diets have led to improved skin symptoms in some, but not all of these patients. Whether food allergy can exacerbate AD is still controversial. A systematic review of nine randomized controlled trials that evaluated dietary exclusions in patients with AD found no effect.

In a study by Sampson and McCaskill, 113 patients with severe AD and food allergy were evaluated with double blind, placebo-controlled food challenge (DBPCFC). Sixty-three children (56%) had positive food challenges; 84% of these children developed skin symptoms; 72% of the reactions noted were attributed to egg, peanut and milk. Approximately 40% of the 40 patients who were re-evaluated after appropriate elimination diets were initiated showed loss in food hypersensitivity after 1 to 2 years, with no obvious positive effect on the AD.

There were two reviews that evaluated the effect of dietary exclusion for treating AD. Kramer and colleagues evaluated maternal dietary avoidance during breast-feeding by mothers of infants with AD. They found no significant reduction in eczema between infants whose mothers avoided dietary allergens and those whose mothers followed a normal diet. Bath-Hextall and colleagues evaluated the effect of dietary exclusion by patients with AD. They evaluated a number of prior studies and found no evidence to support the use of dietary exclusion strategies for treating AD.

Thus, children aged younger than 5 years with significant AD should be evaluated for food allergy, especially to milk, eggs, peanut, wheat and soy. This should especially be done in a child whose AD does not respond to appropriate medical management, and also if the child has a reliable history of having increased eczema after eating.

Patients with documented food allergy who also have AD should avoid that food allergen. Such avoidance may reduce the severity of the AD, but there is not enough evidence whether such avoidance will alter the course of the AD. Before eliminating food such as milk, wheat, soy bean or egg from the diet, it is important to ensure that these children are clinically allergic to the food, either by a good history or a DBPCFC. Remember that many children will lose their clinical sensitivity to these foods in time and should be evaluated yearly by their allergist.

For more information:

• Bath-Hextall F. Allergy. 2009;64;258-264.
• Bath-Hextall F. Cochrane Database Syst Rev. 2008;1:CD005203.
• Boyce JA. J Allergy Clin Immunol. 2010;126:1105-1118.
• Eigenmann PA. Pediatr Allergy Immunol. 2000;11:95-100.
• Eigenmann PA. Pediatrics. 1998;101:E8.
• Forbes LR. Pediatric Annals. 2009;38:84-90.
• Garcia C. Allergol Immunopathol (Madr). 2007;35:15-20.
• Hill DJ. Clin Exp Allergy. 2008;38:161-168.
• Hill DJ. Pediatr Allergy Immunol. 2004;15:421-427.
• Kramer MS. Cochrane Database Syst Rev. 2006;3:CD000133.
• Rona RJ. J Allergy Clin Immunol. 2007;120:638-646.
• Sampson HA. J Pediatr. 1985;107:669-675.
• Sampson HA. J Pediatr. 1989;115:23-27.
• Thompson MM. Dermatol Ther. 2006;19:91-96.
• Zuidmeer L. J Allergy Clin Immunol. 2008;121:1210-1218.

Gary S. Rachelefsky, MD, is clinical professor of allergy and immunology at the Geffen School of Medicine at UCLA and director of the Executive Care Center for Asthma, Allergy & Respiratory Diseases.

Disclosure: Dr. Rachelefsky reports no relevant financial disclosures.