As there are no generally agreed standard methods for the detection of Helicobacter pylori, a standardized protocol for the C-13-urea breath test (C-13-UBT) was proposed and evaluated. A novel 100 ml liquid fatty test meal to maximally delay gastric emptying was taken 10 min before C-13-urea (100 mg in 50 ml of water). Two-litre expired breath samples were collected, one every 5 min, from 10 min (t = 10) up until 40 min (t = 40) (after ingestion of the C-13-urea in water) into a large reservoir collecting bag, from which at the end of the test a single 20 ml sample was taken for analysis by mass spectrometry. In addition, a single point sample was taken directly into a vacutainer 30 min after C-13-urea ingestion (t = 30). The results were compared with antral histology (two biopsies haematoxylin and eosin and Gimenez method), culture (two biopsies, microaerophobic culture using Oxoid SR147), CLO-test and enzyme linked immunosorbent assay (ELISA) serology using double antigen (120 kD and ultracentrifuged cell sonicate) coated wells. The 'gold standard' was defined by the result of any two of the biopsy based methods. H. pylori was detected by the 'gold standard' in 149 out of 195 (76%) patients undergoing routine upper gastrointestinal endoscopy. The C-13-UBT has a specificity of 98%, with a sensitivity of 99%, in comparison with ELISA serology (83%, 95%), CLO-test (100%, 92%), histology (97%, 95%) and culture (98%, 83%), respectively. The new European standard method for the C-13-UBT is a good as, and may be better than, other techniques for the detection of H. pylori.