AIM: The Lewis b (Leb) antigen has been implicated as a possible binding site for attachment of Helicobacter pylori (H pylori)to gastric mucosa. However, studies both supporting and denying this association have been reported in the literature. Differences in secretor (Se)genotype have been suggested as a possible reason for previous discrepancies. Therefore, we investigated the relationship between Le and Se genotypes and H pylori infection rates in people with peptic ulcer or gastric cancer.METHODS: Peripheral blood samples were obtained from 347 patients with endoscopic evidence of peptic ulcer disease (235 cases of duodenal ulcer, 62 of gastric ulcer,and 50 of combined duodenal ulcer/gastric ulcer) and 51patients with gastric cancer on endoscopy. Peripheral blood specimens from 101 unrelated normal volunteers were used as controls. Lewis phenotype was determined using an antibody method, whereas Le and Se genotypes were determined by DNA amplification and restriction enzyme analysis. Gastric or duodenal biopsies taken from patients with endoscopic evidence of peptic ulcer or gastric cancer were cultured for H pylori. Isolates were identified as H pylori by morphology and production of urease and catalase. The H pylori infection status was also evaluated by rapid urease test (CLO test), and urea breath test (13C-UBT). Results of studies were analyzed by chi-square test (taken as significant).RESULTS: H pyloriwas isolated from 83.7% (303/347)of patients with peptic ulcer disease. Statistical analysis did not show any significant difference in Lewis phenotype or genotype between patients with and without H pylori infection. No significant association was found between Lewis genotype and peptic ulcer or gastric cancer.CONCLUSION: Lewis blood genotype or phenotype may not play a role in the pathogenesis of H pyloriinfection.However, bacterial strain differences and the presence of more than one attachment mechanism may limit the value of epidemiological studies in elucidating this matter.