Children's Doctor


Atypical Food Allergy

Terri Brown-Whitehorn, MD

Baby J is a 5-month-old, former full-term infant, referred for evaluation. He was exclusively breast-fed until 4 months of age. His mother’s diet was varied and Baby J was well, without eczema, gastroesophageal reflux, or irritability. At 4 months of age, he was given a little rice cereal without difficulty. A week later, he was given a few bites of apples without difficulty. Seven days before admission, he was given rice cereal and apples. Within an hour, he developed nonbilious emesis (2 hours of dry heaving), became limp, and developed glazed eyes. He was brought to his local hospital and admitted for evaluation. His stools were found to be heme-positive. The workup included a normal CT scan of the head, normal abdominal radiograph, and normal barium enema. He was diagnosed with an apple allergy and discharged home with instructions to avoid apples, and the family received an EpiPen Jr. to use as needed.

For the next few days, he did well exclusively breast-fed. On the day before admission, he was given a spoonful of rice cereal. Within 2 hours, he developed emesis (nonbilious, repetitive). Upon arrival at his local ED, he was limp and ill-appearing. He was afebrile, his heart rate was 157 bpm, and his blood pressure was 63/45. He was given subcutaneous epinephrine without improvement. He was given intravenous fluids and transported to The Children’s Hospital of Philadelphia.

Discussion: Food protein-induced enterocolitis (FPIES) is a condition in which infants and toddlers present with excessive vomiting and diarrhea a few hours after ingestion of a specific food protein (although may be delayed as long as 6 hours after ingestion). This reaction is typically quite severe and may lead to lethargy, hypotension, acidemia, methemoglobinemia, and shock. Symptoms resolve when the offending food is removed and the reaction is treated, and recur when reintroduced. Allergy confirmation requires negative workup for other diseases, including infectious, metabolic, or structural abnormalities.

Little is known about the exact pathophysiology of FPIES, although it is thought to be more of a T-cell driven, non- IgE-mediated reaction than an IgE-mediated reaction. Thus, one often finds negative skin prick testing and specific IgE (in the past called RAST) testing. We have found that atopy patch testing, which looks for delayed reaction to food antigens, has been useful in determining responsible food antigen(s) in many patients (see Figure 1). For those in whom diagnosis is uncertain by history or patch testing, the gold standard for diagnosis remains food allergy challenge specific for the diagnosis of FPIES.

The most common inciting foods include milk, soy, and grains (rice, wheat, barley, and oat) although recent case reports suggest that poultry, sweet potatoes, green beans, and peas have also been culprits. For some infants and toddlers, reaction occurs with very small amounts of food (1 teaspoon of rice cereal, for instance), whereas others may require a full serving. Regardless of the amount, the reaction can be quite severe.

Treatment of an immediate reaction should include intravenous fluids, supportive care, and steroids. Unlike with IgE-mediated food allergy reactions, these children typically do not respond to epinephrine or antihistamines.


Figure 1:This child’s atopy patch test is positive for soy, corn, beef, wheat, egg, and rice

Long-term therapy includes strict avoidance of the offending food in both the child and the breast-feeding mother (if applicable). For those with milk- and soy-induced FPIES (the most common causes), an extensively hydrolyzed casein formula is sometimes tolerated, but we recommend an amino acid-based formula.

Most children will outgrow their sensitivity within 2 years of their last exposure, although persistence of reaction may occur. Repeated patch testing has been useful in some cases to correlate with outgrowing sensitivity. However, even with negative patch testing we often perform FPIES-directed food challenges in the hospital, due to the severity of potential reactions.

Baby J was found to have rice-induced enterocolitis by history and patch testing. Strict avoidance was recommended. Apples were reintroduced to his diet without reaction. At the time of patch testing, other grains were found to be positive (wheat, barley, and oat) and over time, reactivity was confirmed by food challenge. He did require intravenous fluids and steroids, and overnight hospitalizations for these oat and wheat challenges, which were performed at 3 years of age. He has not required readmission to the intensive care unit. Unlike many patients who outgrow their sensitivity by 4 years of age, he must continue strict avoidance of rice, wheat, barley, and oat. He has not had any re-exposure to rice, either accidentally or by food challenge, to date.

References and Suggested Readings
Fogg MI, Brown-Whitehorn T, Pawlowski NA, Spergel JM. Atopy patch testing for the diagnosis of food protein-induced enterocolitis syndrome.
Pediatr Allergy Immunol. 2006;17:351-355.

Sicherer S. Food protein-induced enterocolitis syndrome: case presentations and management lessons.
J Allergy Clin Immunol. 2005; 115:149-156.

Maloney J, Nowak-Wegrzyn A. Educational clinical case series for pediatric allergy and immunology: allergic proctolitis, food protein induced enterocolitis syndrome and allergic eosinophilic gastroenteritis with protein losing enteropathy as manifestations of non-IgE-mediated cow’s milk allergy.
Pediatr Allergy Immunol. 2007;18:360-367.

Hives: the Wheal Truth

James Corry, MD

EG is a 3-year-old child who developed hives 8 days before his visit. His hives begin during the day and fade in a few hours. They are generalized, pruritic, nonbruising, and on 2 occasions were associated with lip edema. They have occurred on 4 separate days. Benadryl relieves pruritis.

SE is a 10-year-old child who has had hives for the last 8 weeks. The hives occur most days of the week, although he can be clear for as long as 36 hours. Many times they begin overnight or on awakening and will clear within 24 hours. They are generalized, pruritic, and nonbruising without angioedema. There are no systemic symptoms. Cetirizine reduces pruritis and severity. Two years ago, a similar episode of hives lasted 2 months.

Discussion: The first case describes a child with acute urticaria (less than 6 weeks’ duration). In such cases, a focused review is likely to reveal a food, medication, environmental exposure, viral infection, or insect sting as a causal factor. Although food allergy in this case is unlikely given the 4 episodes in a short time, food is a common cause of acute urticaria. It is important to concentrate on any foods eaten within 2 hours of onset. The most common foods implicated are milk, egg, fish, peanut, nuts, soy, or wheat. Foods can also cause hives at areas of contact. This is particularly common with strawberry and tomato. Hives caused by food will resolve within 24 hours and not recur unless the food is reingested. Skin tests or specific IgE testing will confirm the presence of sensitivity to a suspected food. Patients found to have a food allergy should have an EpiPen and instructed food avoidance.

The history should also focus on new medications, topical applications, and environmental exposures, such as pets, high pollen counts, or new hobbies. Respiratory symptoms are usually present with aeroallergen exposure. Opioids such as codeine or NSAIDs commonly cause hives through pharmacologic mechanisms. Avoidance of the food or medication is sufficient to control the hives.

Viral illness is a common cause of acute urticaria. Accompanying symptoms will reveal the presence of infection. On occasion, the hives may persist for several weeks.

The second patient has had hives for longer than 6 weeks and would be considered chronic (see Figure 2). It is unusual to discover an allergic trigger in this situation. After a careful history, an etiology is found in only 15% of cases, with physical factors most common. Patients should be questioned about pressure on the skin along with cold, heat, or solar exposure in relation to urtication. Cholinergic urticaria can be recognized by the small papules occurring with exercise and heat. Ice cube challenge will confirm cold urticaria.


Figure 2: A child with chronic urticaria

Although unusual in children, autoimmune disease, vasculitis, and urticaria pigmentosa can present as urticaria. Suspicion is increased with bruising, lesions lasting longer than 24 hours, or systemic symptoms such as joint swelling or pain, fever, lethargy, change in appetite, or weight loss. Subclinical thyroiditis is common in adults and has been reported in children in cases of chronic urticaria.

There is no standard evaluation for chronic urticaria, and investigations should be directed by symptoms, response to therapy, and clinical suspicion. Therapy with nonsedating antihistamines exceeding standard doses will control many patients. Addition of H2-blocking antihistamines can provide added benefit when combined with standard antihistamines. Most children will go into remission in several months, but occasionally urticaria can persist for years.

References and Suggested Readings
Baxi S, Dinakar C. Urticaria and angioedema.
Immunol Allergy Clin North Am. 2005;25:353-367.

Volonakis M, Katsarou-Katsari A, Stratigos J. Etiology factors in childhood chronic urticaria.
Ann Allergy. 1992;69:61-65.

Penicillin Problem

Trong Le, MD

A 5-year-old girl presented to her pediatrician with complaints of a low-grade fever, congestion, headache, and cough at night for 1 1/2 weeks. Streptococcus rapid antigen and culture were negative. At her follow-up visit 5 weeks later, she complained of increased frontal headache and congestion. Her doctor ordered a sinus MRI without contrast and was about to prescribe penicillin, but her mother said her daughter was allergic to it—she had taken it 4 years ago and broke out in an itchy, red rash within 1 hour of her second dose. Instead, she was treated with Zithromax for 5 days. Her head MRI showed complete opacification of the maxillary sinuses. Subsequently, she was treated with cefdinir (Omnicef, a third-generation cephalosporin) for 20 days resulting in only modest clinical improvement.

Discussion: As many as 12% of children treated with an antibiotic experience a cutaneous adverse drug reaction (ADR). However, the majority of people reporting antibiotic allergy are falsely labeled allergic or lose their sensitivity as a result of years of avoidance. Consequently, previous adverse antibiotic reactions without proper evaluation can adversely affect subsequent choice(s) of medical therapy. Such clinical dilemmas may be remedied with an understanding of common drug-associated cutaneous manifestations, use of available diagnostic tools, and timely management.

A small number of drug-related cutaneous reaction patterns are observed in the pediatric population, including exanthematous, urticarial, fixed, photoallergic, serum sickness-like, and exfoliative drug eruptions. Of these, exanthematous and urticarial drug reactions are the most common. Non-IgE-mediated exanthematous drug eruptions typically appear as nonpruritic, maculopapular rashes and usually develop during the first 4 to 8 days of therapy. These ADRs are often incidentally misattributed because many cases in children occur in conjunction with viral infections; however, only about 4% of reported cases are confirmed as true drug allergies. In contrast, early recognition of cutaneous ADRs associated with delayed onset from initial exposure of about 2 weeks (or 2 days with re-exposure to associated high-risk anti-infectives, particularly sulfonamides), may prevent the progression into Stevens-Johnson syndrome or toxic erythroderma necrolysis (TEN). Acute urticarial IgE-mediated reaction usually develops within 24 hours of the onset of therapy or within an hour of exposure following sensitization.

Viral-induced urticaria can also mimic IgE-mediated reactions. Due to the common confounding association of viral cutaneous presentation, a significant number of mislabeled drug-associated rashes may be avoided with judicious prescription of antimicrobials in children who have febrile illness without clear indication.

Evaluation of drug-specific IgE antibodies is often useful for confirming the diagnosis and prediction of future IgE-mediated reactions, such as anaphylaxis and urticaria. Skin tests are preferred, because they have excellent negative predictive values and better sensitivity compared with immunoassays for specific IgE antibodies. Penicillin skin testing is validated and the most reliable in evaluating IgE-mediated allergy. When performed with both major and minor determinants, the negative predictive value of penicillin skin testing for immediate reactions approaches 100%, whereas the positive predictive is not as well studied and ranges between 40% and 100%. Currently, the major determinant (penicilloyl polylysine, Pre-Pen) is not commercially available. In the absence of validated skin test reagents, therapeutic options for patients with a history of penicillin allergy include: (1) prescribing an alternative antibiotic, (2) performing a supervised graded challenge, or (3) performing penicillin desensitization.

Recent studies show that most patients with remote (>3 years), non-life-threatening adverse reactions to penicillin (IgE- or non-IgE-mediated), such as the patient in the case above, are at low risk of developing severe adverse reactions to oral penicillin challenge. Choosing an alternative agent without a referral to an allergist for a complete evaluation of this patient’s status of penicillin sensitivity would result in a poor response to therapy and increased risk for potentially serious clinical complications from inappropriate choices of alternative antibiotics. For instance, the risks of reacting from administering a cephalosporin to a patient with IgE-mediated penicillin allergy are up to 16%, 4%, and about 1% for first-, second-, and third-generation cephalosporins, respectively. A safe and appropriate action for this patient is to refer her to an allergist to undergo skin testing to penicillin G. Because it is not a validated agent and has reduced specificity compared with testing with Pre-Pen, patients with negative skin test results to penicillin G are at increased risk for false negative results and should be submitted to a supervised oral graded challenge with the penicillin compound in question. Less than 3% of those who tolerate the challenge are at risk for resensitization. For patients with a history of anaphylaxis and without an acceptable alternative agent, desensitization is recommended.

References and Suggested Readings
Segal AR, Doherty KM, Leggott J, Zlotoff B. Cutaneous reactions to drugs in children.
Pediatrics. 2007;120(4):e1082-e1096.

Bernstein IL, Li JT, Bernstein DI, et al. Allergy diagnostic testing: an updated practice parameter.
Ann Allergy Asthma Immunol. 2008;100(3 suppl 3):S1-148.

Residents’ Corner

Rushani Weerasooriya Saltzman, MD

An Itchy Rash

MS is an 18-month-old, ex-full-term female with atopic dermatitis (AD) who presented with vesicles and pustules. They were initially on her chin and rapidly spread to the rest of her body over 2 days. The rash was itchy and she was given antihistamines. She was febrile to 103.7°F the day before presentation. She had decreased oral intake. Her only new exposure had been to horses while visiting a ranch.

Her vitals were significant for T 102.4°F and HR 172. She had a widespread eczematous eruption on her chest and back. Grouped pustules and vesicles and diffuse erythema were noted. She had multiple crusted lesions on her face, including some in the periocular distribution.

A CBC (WBC 17.8 with 58% segmented neutrophils and 32% lymphocytes), CMP (within normal limits), blood and wound cultures, and viral studies (EBV, CMV, adenovirus, HHV-6, parvovirus, HSV, VZV) were sent. She was started on IV acyclovir and vancomycin.

The differential diagnosis for rash and fever in a nontoxic child is extensive, including many viral exanthems. However, in a patient with AD, one should consider eczema herpeticum (EH).

Her exam was consistent with impetiginized AD. Her HSV PCR was positive. She transitioned to oral acyclovir and Bactrim (MRSA-positive wound culture, sensitive to Bactrim). Mupirocin was applied to the pustules and Aquaphor was used for daily moisturizing therapy. An ophthalmologic evaluation revealed no ocular disease. Her rash resolved and she was sent home on 14 days of oral acyclovir and Bactrim. Alclometasone and hydrocortisone 2.5% ointment were prescribed twice daily for the affected lesions.

EH, or Kaposi varicelliform eruption, is an acute, disseminated HSV infection in AD patients. The incubation period can range from a few days to 2 weeks. Skin manifestations include the eruption of multiple, pruritic vesicles and pustules, involving both healthy and eczematous skin.

The pathogenesis of EH involves IL-4, a T-helper cell inhibitor. It causes local reduction of the cell-mediated immune response against HSV, resulting in viral proliferation. Scratching affected skin leads to indirect inoculation of unaffected skin areas. Systemic acyclovir therapy is used to treat EH. Appropriate antibiotic therapy is necessary when bacterial superinfection occurs, primarily targeting S. aureus.


  • Print
  • Share

Contact Us

Send us your feedback and alumni news, or request to receive Children’s Doctor electronically.