The direction of the relationship between Helicobacter pylori and duodenal ulcer

Prevalence of H pylori was determined in 208 patients undergoing endoscopy for dyspepsia. Results were expressed separately for methods involving and not involving polymerase chain reaction (PCR). The results confirmed well-established findings such as a greater prevalence in duodenal ulcer patients than in groups with other diagnoses, and increasing prevalence with age. The unselected records of 320 different patients undergoing endoscopy for dyspepsia were studied using logistic regression analysis to determine factors significantly associated with a) DU and b) H pylori. It was found that number of endoscopies was significant for both regressions. In order to avoid the compounding effects of repeated endoscopies, a comparison was made in patients undergoing endoscopy for the first time in the series between those who had and those who had not undergone a previous endoscopy. This finding indicated that the relationship between prevalence and number of endoscopy was spurious. A search for the explanation revealed that the number of endoscopy was related to the duration of duodenal ulcer symptoms. Applying this information, it was found that 5 patients with symptom of 6 months were free of the infection, but the prevalence after 6 months was 96%. It is concluded that H pylori is most unlikely to be a prime factor in the aetiology of duodenal ulcer