Date Published: September 22, 2017
Publisher: BioMed Central
Author(s): Elissa M. Abrams, Allan B. Becker.
Oral food challenges are the clinical standard for diagnosis of food allergy. Little data exist on predictors of oral challenge failure and reaction severity.
A retrospective chart review was done on all pediatric patients who had oral food challenges in a tertiary care pediatric allergy clinic from 2008 to 2010.
313 oral challenges were performed, of which the majority were to peanut (105), egg (71), milk (41) and tree nuts (29). There were 104 (33%) oral challenge failures. Children were more likely to fail an oral challenge if they were older (P = .04), had asthma (P = .001) or had atopic dermatitis (P = .03). Risk of challenge failure was significantly different between food allergens, with more failures noted for peanut than for tree nuts, milk or egg (P = .001). Among challenge failures, 19% met criteria for anaphylaxis. Significantly more tree nut and peanut challenges met criteria for anaphylaxis than milk or egg (P < .001). Skin test size and specific IgE level were significantly higher in those who failed oral challenges (P < .001). The highest rate of challenge failure and severity of failure was to cashew, with 63% of cashew challenges reacting, of which 80% met clinical criteria for anaphylaxis. The risk of challenge failure differed with type of food studied, with peanut and tree nut having a higher risk of challenge failure and anaphylaxis. Cashew in particular carried a high risk and caution must be exercised when performing these types of oral challenges in children.
Food allergy affects 2–10% of the population, and is more common in children than adults . The diagnosis of food allergy is often based on results of a careful history, skin prick testing (SPT) and serum food-specific IgE . Oral food challenges (OFCs) assist in the diagnosis of food allergy, and are essential to determine whether an allergy has been outgrown . However, OFCs do carry the risk of a systemic allergic reaction . While the double blind placebo controlled food challenge is the most accurate and a true ‘gold standard’ for diagnosis of food allergy, it is time consuming and costly . The open oral food challenge is often used instead, although it is subject to patient bias .
A retrospective chart review was performed on all open oral food challenges between January 1, 2008 and December 31, 2010 at the University of Manitoba pediatric allergy clinic. Oral food challenges were performed based on the clinical decision of the attending physician, with consideration of clinical history, results of epicutaneous testing, and/or results of serum food-specific IgE. Challenges were performed to confirm food allergy, or when there was a suspicion of oral tolerance after a period of avoidance in a food allergic child.
There were 313 oral food challenges performed between January 1, 2008 and December 31, 2010 at the University of Manitoba Pediatric Allergy Clinic. There were 105 peanut, 71 egg, 41 milk, 29 tree nut (6 almond, 1 brazil nut, 8 cashew, 6 hazelnut, 1 macadamia nut, 2 pecan, 5 walnut), 10 finned fish, 14 shellfish, 9 soy, and 34 other challenges performed. Seventeen patients underwent oral challenges to more than one food during this time (although never more than one food each day), and eleven patients had more than one oral challenge to the same food. Some peanut and tree nut challenges were masked (often in pudding).
Our study shares some findings that are similar to previous studies. Oral challenge failure rate of 33% is in keeping with other studies that have reported challenge failure rates varying from 18.8 to 43% [4–10]. Similar to other studies, we found increased risk of challenge failure in children with asthma and eczema. Perry et al’s retrospective review of 604 oral challenges also noted increased risk in children with eczema or asthma, but not other atopic disease outcomes . Our population, similar to Perry et al’s study, is that of a tertiary care facility which may lead to higher atopic rates than seen in other primary or secondary care settings. Finally, similar to previous studies, we found that skin test sizes and serum food-specific IgE levels were significantly higher for failed than passed oral challenges [6, 9, 11].