Skin testing with raw egg does not predict tolerance to baked egg in egg-allergic children.

BACKGROUND: Most children with egg allergy tolerate egg in baked foods, such as cake, but tolerance cannot be predicted with conventional allergy testing. We hypothesized that the skin prick test (SPT) wheal to unprocessed raw egg might predict tolerance of baked egg at formal oral food challenge (OFC). METHODS: We conducted a retrospective chart review to assess the utility of SPT wheal to egg extract (EE), raw egg (RE), and the ratio of EE:RE in predicting outcome of baked-egg OFC in children presenting to our tertiary referral centers with a physician diagnosis of egg allergy and following complete egg avoidance in their diet, between 2009 and 2013. OFC were performed following a standardized protocol using baked egg in cake, to a total dose equivalent to 3g egg protein. RESULTS: Data were analyzed from 186 completed challenges: OFC was positive in 64 (34%) children and negative in 122 (66%). Six children experienced anaphylaxis at OFC. Children tolerant to baked egg were more likely to have a lower SPT to egg extract/raw egg and EE:RE (median 0.56) than their allergic counterparts (0.70, p < 0.05). However, ROC curve analysis demonstrated poor predictivity of challenge outcome, with AUC for SPT to egg extract, raw egg and EE:ER equal to 0.71, 0.63 and 0.60, respectively. CONCLUSION: EE:RE was not helpful in predicting outcome of baked-egg OFC. Indeed, SPT to egg extract was slightly better at predicting outcome than either SPT to raw egg or EE:RE. Unfortunately, tolerance to baked egg can only be predicted from previous history or through controlled exposure.