Methods: Forty-four children suspected to have fruit, nut and/or legume allergy were selected. Patients were classified as allergic if they had presented at least 2 reactions unequivocally related to HZNT ingestion in the last 2 years. Patients were defined as tolerant if they consumed HZNT on a regular basis. Clinical questionnaire, skin prick test (SPT), serum total and specific IgE and MIA-ISAC IgE (Thermo Fisher Scientific, Uppsala, Sweden) were performed.
Results: Sixteen patients (11 males) were defined as allergic and 28 (15 males) tolerant. HZNT-SPT wheal size (mm) (median 7.75;IQR:4-12 vs.2.5; IQR:0-9.5,p=0.000) and HZNT-sIgE (kU/L) (median 14.45;IQR:1.98-370 vs. 0.82;IQR:0.02-14.3,p=0.000) were significantly greater in allergic than in tolerant children. Both positive Cor a9-sIgE and Cor a14-sIgE were significantly more frequent in allergic patients (75.00% vs. 14.28%,p=0.000 and 75.00% vs. 10.71%,p=0.000, respectively). Cor a9-sIgE values were significantly higher in allergic children whether by means of ImmunoCAP (median 4.38 kU/L; IQR:0.26-16 vs. 0.02 kU/L; IQR:0-0,21, p=0.000) or MIA-ISAC (median 0.14 ISU;IQR:0-2.2 vs. 0 ISU;IQR:0-0, p=0.000), as were Cor a14-sIgE values (median 4.97 kU/L;IQR:0.39-20.4 vs. 0.02 kU/L;IQR:0.01-0.09,p=0.000). This was not found for Cor a8. ROC curves were constructed for the three allergens showing Cor a14 the best diagnostic performance (AUC:0.925,95%CI:0.847-1,p=0.000).
Conclusions: Cor a14 is the best discriminating allergen in the diagnosis of HZNT allergic patients.