The aim of the present study was to correlate molecular evidence of the presence of Helicobacter pylori in gastric biopsy samples, based on analysis of 16S rDNA, vacuolating toxin (vacA), urease A (ureA) and cagA genes, with the clinical, histological and serological findings in patients with H. pylori-associated gastritis. Fresh biopsy samples were collected from the gastric antrum and corpus of 22 asymptomatic volunteers with or without H. pylori-associated gastritis. Total DNA was extracted from the biopsy material and subjected to 16S rDNA PCR amplification, Southern blotting and 16S rDNA sequence analysis of the PCR products. The vacA, ureA and cagA genes were characterised by PCR amplification and Southern blot analysis. Based on partial 16S rDNA sequence analysis, DNA belonging to the genus Helicobacter was detected in gastric biopsy samples from 20 of 22 subjects, including seven of nine histologically and serologically normal controls. Six of 20 partial 16S rDNA sequences revealed variations within variable regions V3 and V4 that deviated from those of the H. pylori type strain ATCC 4350T and, therefore, possibly represented other species of Helicobacter. VacA genes identical with those of the type strain were found predominantly in the subjects with H. pylori gastritis, and all the patients except one were found to be cagA-positive. There was no evidence of false positive PCR reactions. In conclusion, the PCR-based molecular typing methods used here were apparently too sensitive when applied to the detection of H. pylori in human gastric tissues. The lack of quantitative analysis makes them inappropriate as clinical tools for the diagnosis of H. pylori-associated gastritis, despite the fact that they provide a qualitative and sensitive tool for the detection and characterisation of H. pylori in the gastrointestinal tract.