H1-antihistamines and urticaria: how can we predict the best drug for our patient?



Martin K Church, Department of Dermatology and Allergy, Charité – Universitätsmedizin Berlin, Charitéplatz 1, D-10117 Berlin, Germany.

E-mail: mkc@southampton.ac.uk


Urticaria, and especially chronic spontaneous urticaria (CSU), is a difficult condition to treat. Consequently, clinicians need to use the best H1-antihistamines currently available and the pharmaceutical industries need to keep developing H1-antihistamines that are more effective than the ones we have today. To do this we need to be able to compare the clinical efficacy of both established and new drugs. Obviously, the ideal way to do this is to use head-to-head studies in CSU. However, such studies are extremely expensive and, in the case of novel molecules, have ethical and logistical problems. Consequently, we need to have predictive models. Although determination of Ki, an indicator of the in vitro potency of an H1-antihistamine, may help in the initial selection of candidate molecules, the large differences in volume of distribution and tissue accumulation in humans, precludes this from being a good predictor of clinical efficacy in CSU. From the data reviewed in this article, especially the direct comparative data of desloratadine and levocetirizine in weal and flare studies and CSU, weal and flare response would appear to be the best indicator we have of effectiveness of H1-antihistamines in clinical practice. However, it must be pointed out that the conclusion is, essentially, based on detailed comparisons of two drugs in studies sponsored by pharmaceutical companies. Consequently, to confirm the conclusions of this review, a multicentre study independent from the influence of pharmaceutical companies should be commissioned to compare the speed of onset and effectiveness of desloratadine, fexofenadine and levocetirizine in chronic spontaneous urticaria and against histamine-induced weal and flare responses in the same patients so that we have a clear understanding of the predictive value of our models.