Identifying patients at risk of severe allergic reactions and/or low threshold of reactivity is very important, particularly for staple foods like egg.

BAT was the best biomarker to predict severity and threshold of allergic reactions to BE and can be useful when making decisions about management of egg allergy.

One hundred and fifty children underwent double-blind placebo-controlled food challenge (DBPCFC) to baked egg (BE), skin prick testing and blood collection for serology and basophil activation test (BAT). Patients who passed BE DBPCFC underwent loosely cooked egg (LCE) DBPCFC. Severity of allergic reactions was classified following Practall guidelines and threshold dose was determined during DBPCFC.

Sixty out of 150 (40%) children reacted to BE and 16 out of 77 (21%) to LCE on DBPCFC. Considering DBPCFC to BE, 23 children (38%) had severe reactions and 33 (55%) reacted to 0.13 g or less of egg protein (low threshold group). Two children (2 out of 16 = 12%) had severe reactions to LCE. Demographic, clinical and most immunological features were not significantly different between severe/non-severe BE reactors or low/high threshold groups. Severe BE reactors had higher ovomucoid-sIgE (p = .009) and higher BAT to BE (p = .001). Patients with lower threshold to BE had higher IgE-specific activity (p = .027) and BAT to egg (p = .007) but lower severity score (p = .008). Optimal cut-offs for ovomucoid-sIgE had 100% sensitivity, 35% specificity and 60% accuracy and for BAT 76% sensitivity, 74% specificity and 75% accuracy to identify BE severe reactors. Optimal cut-offs for specific activity had 70% sensitivity, 68% specificity and 69% accuracy and for BAT 70% sensitivity, 72% specificity and 71% accuracy to identify low threshold patients. Conclusions: BAT was the best biomarker to predict severity and threshold of allergic reactions to BE and can be useful when making decisions about management of egg allergy.