Adult IgE-mediated food allergy: clinical characteristics, predictors of severe reactions, and total IgE cut-off

Abstract

Background: Adult immunoglobulin E (IgE)-mediated food allergy (FA) is increasingly recognized, yet it has remained under-characterized, compared with pediatric FA. We described clinical features, diagnostic profiles, predictors of severe reactions and explored a pragmatic total IgE cut-off in an adult cohort.

Method: We retrospectively reviewed 423 consecutive adults evaluated for suspected IgE-mediated FA at a tertiary center. Demographics, comorbid atopy, index culprit foods, diagnostic testing (skin prick/prick-to-prick, serum-specific IgE, component-resolved diagnostics), laboratory parameters, reaction severity (Brown grade severity), and adrenaline autoinjector (AAI) prescribing were extracted. Group comparisons used χ²/Fisher’s exact, t-test, or Mann–Whitney U tests. Receiver operating characteristic analysis assessed the discriminative value of total IgE for confirmed FA. Variables associated with Brown grade 3 severity were examined by multivariable logistic regression.

Results: FA was confirmed in 79/423 patients (18.7%). Median age was 39.2 years; 71.4% were females. Frequent implicated groups were meat (23.9%), fruit (23.4%), nuts (22.5%), and vegetables (19.4%). Atopy was present in 46.1%; sensitization to mites and pollens occurred in 38.3% and 25.1%, respectively. Compared to non-confirmed cases, confirmed FA showed higher proportions of asthma, non-steroidal anti-inflammatory drug allergy, moderate/severe reactions (Brown grades 2–3 severity), atopy, mite/pollen sensitization, latex–fruit and pollen–fruit syndromes, AAI prescription, and food-dependent exercise-induced anaphylaxis (all P < 0.05). Eosinophil count/percentage and total IgE were also high (all P < 0.05). Total IgE ≥110.5 IU/mL predicted confirmed FA with 64.6% sensitivity and 59.7% specificity (area under curve 0.634; 95% confidence interval [CI]: 0.561–0.706; P = 0.001). In multivariable analysis, the absence of rhinitis and fish/seafood allergy, and the presence of atopy were independently associated with Brown grade 3 severity (model accuracy 84.4%). Overall, AAIs were prescribed in 38.1%, more often when FA was confirmed (66.7% vs 31.7%).

Conclusions: Adult FA in a real-world tertiary cohort is clinically heterogeneous and clusters with respiratory atopy. A total IgE threshold of 110.5 IU/mL offers modest discrimination and should complement, not replace, history and allergen-specific testing. Rhinitis, fish/seafood allergy, and lack of atopy identify patients at higher risk of severe reactions and may guide risk stratification and AAI prescribing.