The gastric mucosal distribution of azithromycin, the prototype of a new class of macrolide antibiotics named azalides, was studied in patients with duodenal ulcer and Helicobacter pylori-related gastritis. The time course of ulcer healing, H. pylori infection, and gastritis activity was also evaluated. Twenty patients (median age 50 years) received the following treatment for 1 month: three cycles of azithromycin (500 mg/day for 3 consecutive days) on days 1-3, 11-13 and 21-23 plus omeprazole (40 mg/day) for 30 consecutive days. Endoscopic biopsy specimens of gastric mucosa and blood samples were collected on days 0, 4, 7, 10, 20 and 30. An additional follow-up endoscopy was carried out on day 60. H. pylori infection was determined by both histology and rapid urease test. Azithromycin concentrations in both plasma and gastric mucosa were measured by a microbiological plate assay, using Micrococcus luteus NCTC 8440 as the reference organism. Azithromycin concentrations in plasma ranged between 0.17 mg/L (95% CI: 0.08-0.26; n = 5) and 0.32 mg/L (95% CI: 0.21-0.43; n = 5) throughout the treatment period. In addition, azithromycin concentrations in gastric mucosa were significantly higher than plasma concentrations at all times examined and ranged from 18.5 mg/kg (95% CI: 15-20; n = 20) to 24.6 mg/kg (95% CI: 16.8-32.4; n = 5), Indicating that the drug was highly retained in the target tissue. Accordingly, the ratio of azithromycin mucosal level to plasma concentration varied between 77.9 (95% CI: 56.5-99.3; n = 5) and 112.7 (95% CI: 100.2-125.2; n = 5). At the end of treatment (day 30) H. pylori was no longer detected in 16 of 20 patients (80%), and this finding was consistent with a marked decrease in the grading of gastritis activity. At the follow-up endoscopy (day 60) the infection was eradicated in only four patients (20%). These data indicate a favourable distribution of azithromycin into gastric mucosa of patients with H. pylori infection and suggest that this new macrolide antibiotic represents a valuable option for treatment regimens against H. pylori. However, the low eradication rate achieved with azithromycin plus omeprazole is a source of concern and requires further investigation.