The results of the in vitro metronidazole resistance on Helicobacter pylori (H. pylori) eradication have been inconclusive. Metronidazole resistance varies among different geographical locations and a previous study from Thailand reported an in vitro metronidazole resistance of H. pylori of 51 per cent. This study was designed to investigate further the effect of the in vitro metronidazole resistance on the outcome of eradication of H. pylori in the Thai population. Fifty two patients with active gastric ulcer (GU) and duodenal ulcer (DU) who had positive culture for H. pylori were studied. All of these patients had positive rapid urease test (CLO test, Delta West, Australia) using gastric biopsy specimens from the antrum and body taken at the time of initial upper endoscopy. In vitro antimicrobial susceptibility test was performed using Epsilometer test (AB Biodisk, Solna, Sweden). All patients received a one-week triple regimen consisting of omeprazole 20 mg twice daily, clarithromycin 500 mg twice daily, metronidazole 500 mg twice daily. Patients with GU continued with another five weeks of omeprazole 20 mg twice daily and patients with DU received another three weeks of omeprazole 20 mg twice daily. Upper endoscopy was repeated at four weeks after the end of the treatment. Three antral and two body biopsy specimens were obtained for identification of H. pylori using CLO test, histology (modified Giemsa stain) and culture. All of these tests had to be negative to confirm a successful eradication. Metronidazole-resistant (MR) strains with MIC > or = 32 mg/l were identified in 27 of the 52 patients (51.92%), whereas, metronidazole-susceptible (MS) strains were isolated from 25 patients (48.08%). Five patients were lost to follow-up and one patient had drug allergy. Successful eradication as defined by negative CLO test, histology and culture was attained in 17/23 (73.91%) patients (GU = 6, DU = 16, GU and DU = 1) with MR strains. 20 out of 23 (86.96%) patients (GU = 9, DU = 12 GU and DU = 2) who had MS strains. The difference was not statistically significant in both groups (P > 0.05). The ulcer healing was, however, highly achieved in both groups (MS = 95.65%, MR = 91.30%, P > 0.05). In vitro metronidazole resistance was high in this population group although this does not predict the outcome of eradication in patients with GU and DU.