What is the most appropriate investigation for a child with suspected atopic dermatitis who has not responded to topical corticosteroids and is suspected of having a food allergy?

Investigation for Suspected Food Allergy in Atopic Dermatitis Unresponsive to Topical Corticosteroids

The most appropriate investigation is limited food-specific IgE testing for common allergens (milk, eggs, wheat, soy, and peanuts), not broad panel testing, and only if the child is under 5 years old with moderate to severe disease that has truly failed optimized topical therapy. 1, 2

Critical First Step: Optimize Treatment Before Testing

Before pursuing any food allergy investigation, you must ensure the atopic dermatitis has been adequately treated, as this is a common pitfall. 1

• Verify that appropriate-potency topical corticosteroids were actually used - many cases labeled as "treatment failure" reflect inadequate therapy rather than food allergy. 1
• Ensure proper moisturizer use and skin care regimen - the foundation of AD management must be optimized first. 1
• Food allergy is only a contributing factor in approximately 35% of children with moderate to severe atopic dermatitis, so most cases will not have a food trigger. 1

When to Proceed with Testing

Limited food allergy testing is appropriate only when ALL of the following criteria are met: 1, 2

• Child is less than 5 years old 1
• Moderate to severe atopic dermatitis (not mild disease) 1
• Persistent disease despite truly optimized topical therapy and skin care 1, 2
• OR a reliable history of immediate allergic reaction (hives, respiratory symptoms, vomiting within 2 hours) after ingestion of a specific food 1

The Correct Investigation: Limited Specific IgE Testing

Test only for the five most common pediatric food allergens: cow's milk, eggs, wheat, soy, and peanuts. 1, 2

Why This Approach:

• Broad panel allergy testing without clinical history is explicitly not recommended - it leads to false-positive results that reflect sensitization rather than true allergy. 1
• Specific IgE testing has high negative predictive value (>95%) but low positive predictive value (40-60%), meaning negative tests effectively rule out allergy, but positive tests require clinical correlation. 1
• Skin prick testing is an alternative, particularly useful if the child has extensive eczematous lesions, prominent dermatographism, or recent antihistamine use. 1

Why the Other Options Are Incorrect

Option A (IgE levels for common food allergies):

This is partially correct but requires the important caveat that testing should be limited to the five common allergens listed above, not broad panels. 1, 2

Option B (PPD skin test):

This is completely irrelevant - PPD tests for tuberculosis exposure and has no role in food allergy or atopic dermatitis evaluation. [@General Medicine Knowledge@]

Option C (Continue food with supervision):

This is dangerous without proper evaluation - if true IgE-mediated food allergy exists, continued exposure could lead to anaphylaxis. 1

Critical Next Steps After Testing

• Positive tests must be confirmed with supervised oral food challenges - the double-blind, placebo-controlled food challenge (DBPCFC) is the gold standard. 1, 2
• Never implement elimination diets based solely on positive IgE tests - this leads to unnecessary dietary restrictions, nutritional deficiencies, growth delay, and calcium deficiency. 1, 2
• Refer to an allergist for comprehensive evaluation and management when food allergy is suspected in the context of persistent moderate-to-severe AD. 1

Common Pitfalls to Avoid

• Do not order total IgE levels - these are not helpful in determining food allergy. 1
• Do not test without a clear clinical indication - testing asymptomatic children or those with well-controlled AD leads to overdiagnosis. 1, 2
• Do not forget that effective AD treatment remains centered on skin care and topical therapies, even when food allergy is confirmed. 1, 2
• Remember that most children develop tolerance to milk, egg, soy, and wheat over time, so periodic re-testing is necessary. 1, 2