Abstract

BACKGROUND: Patients sensitized to birch pollen frequently suffer from a food allergy to plant foods such as celery, carrots, or hazelnut. One of the main manifestations of birch pollen-related food allergy is the oral allergy syndrome. Skin tests and allergen-specific immunoglobulin (Ig) E determinations are poor predictors of such reactions when assessed by double-blind placebo-controlled food challenge (DBPCFC). OBJECTIVE: To investigate whether a cellular test based on leukotriene release from basophils, the cellular antigen stimulation test in combination with enzyme-linked immunosorbent assay (CAST-ELISA), is predictive of pollen-related food allergy. METHODS: Birch pollen-sensitized patients with positive DBPCFC to celery (n=21), hazelnut (n=15), and carrot (n=7) underwent skin tests along with determination of specific IgE and CAST-ELISA for the respective allergens. The results were compared with those of 24 birch pollen-sensitized patients with negative open food challenge to celery, hazelnut, and carrot. RESULTS: While skin prick tests had a sensitivity of 85%, 80%, and 29% for commercial extracts of celery, hazelnut, and carrot, respectively, prick testing with self-prepared extracts yielded sensitivities of 100%, 80%, and 100%, respectively. For specific IgE determinations, sensitivities were 71%, 73%, and 57%, respectively, and the respective specificities were 67%, 73%, and 60%. For CAST-ELISA with various sources and doses of allergens, the sensitivity varied from 71% to 95% for celery, 73% to 80% for hazelnut, and 43% to 86% for carrot. The respective specificities were 67% to 92%, 75% to 88%, and 77% to 91%. Analysis of the predictive value of CAST-ELISA with receiver operating characteristic curves showed that the results of the tests were more predictive of pollen-related food allergy than quantitative allergen-specific IgE determinations. CONCLUSIONS: CAST-ELISA is more specific than routine diagnostic tests for the diagnosis of pollen-related food allergy to celery, hazelnut, and carrot.