RESUMO
BACKGROUND: Immediate hypersensitivity reactions to clavulanic acid (CLV) seem to be on the increase. Diagnosis is mainly based on skin testing and the drug provocation test (DPT), procedures that are not risk free. The aim of this study was to evaluate whether the histamine release test (HRT) could help evaluate patients with selective hypersensitivity to CLV. METHODS: Eighteen patients with immediate selective hypersensitivity reactions to CLV (positive skin tests to CLV but negative to the major and minor determinants of benzylpenicillin and amoxicillin; negative DPT to benzylpenicillin and amoxicillin) and 21 controls with tolerance to CLV were included. Direct and passive HRT, using patient whole blood or 'IgE-stripped' donor blood sensitized by patient serum, respectively, were performed by stimulating the blood with CLV, and basophil histamine release was detected by fluorometric determination. RESULTS: The clinical symptoms were anaphylaxis (n = 6), urticaria (n = 9) and urticaria-angioedema (n = 3). The median time interval between the reaction and the study was 225 days (interquartile range, IQR: 120-387.5) and between drug intake and the development of symptoms 30 min (IQR: 6.25-30). We obtained similar data for both the direct and passive HRT, with a sensitivity and specificity of 55 and 85%, respectively, a positive predictive value of 76% and a negative predictive value of 69%. CONCLUSIONS: The sensitivity of both the direct and passive HRT for diagnosing patients with immediate allergy to CLV is less than 60%. However, the passive HRT has the advantage that it is based on the testing of serum samples that can be handled more easily than fresh blood samples.