Previous investigations have suggested that nasal secretions, obtained by lavage or scraping, and the nasal submucosa, sampled by biopsy, are two distinct compartments. We investigated the effect of intranasal corticosteroids on antigen-induced eosinophil influx into both compartments. We performed a double-blind, placebo-controlled study in 15 patients with seasonal allergic rhinitis. Beclomethasone dipropionate, 84 microg twice a day, was delivered to one nostril while the other nostril received placebo for 1 wk. Subjects were then challenged with grass or ragweed extracts on each inferior turbinate. Nasal scrapings from both inferior turbinates were obtained before and 24 h after challenge, and bilateral inferior turbinate biopsies were obtained 24 h after challenge, with the subjects still receiving treatment. Intranasal steroids led to a significant reduction in sneezes and eosinophil influx in nasal secretions without affecting the number of eosinophils in the submucosa. Furthermore, intranasal steroids had no effect on the numbers of submucosal EG2+ (activated eosinophils) or CD25+ (IL-2-receptor-bearing) cells, nor did they decrease the endothelial expression of vascular cell adhesion molecule-1 (VCAM-1). These data show that pretreatment with intranasal steroids successfully inhibited the clinical response to allergen and reduced eosinophils in the superficial compartment of the nasal mucosa, but it had no effect on inflammation in the deeper compartment. This might be related to a different distribution of the active medication and antigen into the nasal mucosa or to a specific effect of the active medication on the epithelium resulting in inhibited migration of eosinophils across this layer.