Ortolani, C., B.K. Ballmer-Weber, K.S. Hansen, M. Ispano, B. Wüthrich, C. Bindslev-Jensen, R. Ansaloni, L. Vannucci, V. Pravettoni, J. Scibilia, L.K. Poulsen, E.A. Pastorello, 2000. Hazelnut allergy: A double-blind, placebo-controlled food challenge multicenter study. J Allergy Clin Immunol. 105:577-581.
Background: Tree nuts are a common cause of food allergy in Europe. However, few studies deal with real food allergy to hazelnuts in subjects believed to be allergic to this food. Objective: We sought to select subjects with a history of allergic reactions on ingestion of hazelnut and determine how many of these have true allergy by means of the double-blind, placebo-controlled food challenge (DBPCFC). Methods: Eighty-six subjects with a history of symptoms after hazelnut ingestion were recruited from 3 allergy centers (Milan, Zurich, and Copenhagen). All subjects underwent skin prick tests (SPTs) with aeroallergens and hazelnut, as well as having their specific hazelnut IgE levels determined. Diagnosis of clinical relevant food allergy was made on the basis of the DBPCFC. Results: Sixty-seven (77.9%) of 86 subjects had a positive DBPCFC result; 8 were placebo responders, and 11 were nonresponders. Of the 11 nonresponders, 4 had positive open-challenge test results. Of the DBPCFC-positive subjects, 87% also had positive skin test responses to birch pollen extract. Specific IgE determination for hazelnut (positive CAP response ≥0.7 kU/L [ie, class 2]) showed a sensitivity of 0.75, a positive predictive value (PPV) of 0.92, a specificity of 0.16, and a negative predictive value (NPV) of 0.05. Skin tests with commercial hazelnut extract produced a sensitivity of 0.89, a PPV of 0.92, a specificity of 0.05, and an NPV of 0.05. Skin tests with natural food produced a sensitivity of 0.88, a PPV of 0.94, a specificity of 0.27, and an NPV of 0.15. Conclusion: This study shows that hazelnut is an allergenic source that can cause real food allergy, as confirmed by DBPCFC. Skin and IgE tests demonstrated reasonable sensitivity and PPV but a very low specificity and NPV, thus implying that these should not be used to validate the diagnosis of food allergy to hazelnut.