Abstract
Early detection of CSU patients with low probability of a clinical response with antihistamines could undergo prompt initiation of therapeutic alternatives. The aim of the study was to develop and internally validate a model for predicting the clinical response to antihistamines in adult patients with chronic spontaneous urticaria (CSU), who consult allergology and dermatology care centers. A cohort of CSU patients, recruited from four participating centers, were followed up for 12 months. Fifteen candidate variables were selected to be included in the multivariate model and then internal validation was done with bootstrap analysis with 1000 simulations. The outcome variable, clinical response to antihistamines, was evaluated with the UAS (Urticaria Activity Score) scale for seven days: "No response to antihistamines" was defined as UAS7 ≥7 points after at least one month with a maximum dose of antihistamines, while "Response to antiH1" was defined as UAS7 ≤6 points for at least three months with the use of antiH1. A total of 790 patients were included. Among the different models analyzed, the model that included age, angioedema, anxiety/depression, time with the disease, NSAIDs (Non-steroidal anti-inflammatory drugs) intolerance, and UAS7 baseline was considered the one with the best performance (accuracy 0.675, HL 0.87, AUC 0.727). The internal validation analyses demonstrated good consistency of the model. In conclusion, this prediction model identifies the probability of response to antihistamines in patients with chronic spontaneous urticaria. The model could be useful for a personalized therapeutic approach according to individual patient risk.