(1) H pylori is involved in the pathogenesis of GORD so that eradication of this organism would lead to an improvement in, or even elimination of, GORD

H pylori infection interferes with the regulation of gastric acid secretion. The effect of eradication therapy on acid secretion is closely linked to the type of gastritis before treatment. In the case of duodenal ulcer patients with antrum dominant gastritis, acid hypersecretion normalises within some months, while in the case of gastric ulcer patients with more severe corpus gastritis, acid hyposecretion disappears within a few weeks.1 It is conceivable, although not yet unequivocally proved in patients with confirmed GORD, that treatment of H pylorican lead to an improvement in, or the healing of, GORD (at least in individual cases). To date, the literature contains two well designed clinical studies on the treatment of H pylori in patients with GORD; one study was unable to show any influence of eradication on the natural course of GORD over a period of one year and the other study suggested a benefit of eradication therapy.2 ,3 In addition, patients with peptic ulcer frequently experience an improvement or even resolution of heartburn with eradication of H pylori and healing of ulceration.4 Nevertheless, a major role ofH pylori infection in the pathogenesis of GORD is unlikely.5 This evaluation is based on epidemiological time trends, case control studies, and some but not all eradication studies, suggesting a protective rather than an aggravating effect of the infection with respect to GORD.

(2) Eradication of H pylori infection is associated with an improvement in the effectiveness of treatment with PPIs in patients with GORD

H pylori increases the pH elevating effect of PPIs.6 The ammonia generated byH pylori,7 and probably also aggravation of corpus gastritis under PPIs, play a decisive role in this phenomenon.8 Presumably due to this effect, short term studies suggested that PPIs work better in patients with GORD andH pylori infection than in uninfected GORD patients.9 In a multivariate analysis of long term omeprazole studies, H pylori proved to be a predictor of the success of treatment10 while long term cohort studies argue against a clinically relevant effect ofH pylori with respect to the required maintenance dose of omeprazole.11 ,12

(3) PPI treatment of GORD carries a higher risk in patients withH pylori infection than inH pylori negative patients, and this risk can be eliminated by eradication of the infection

Numerous studies have, without exception, shown that inH pylori infected patients receiving PPI treatment, the severity of corpus gastritis in terms of chronicity and activity increases, while in the antrum gastritis improves.13 It may be assumed that the increase in gastritis severity also leads to an increase in mutagenic risk. The extent to which the shift in gastritis per se actually leads to an increase in the risk of gastric cancer remains unclear. Interestingly, the resulting corpus dominant type of gastritis is found disproportionately more frequently in patients with early gastric cancer—as has been shown in a large case control study14—and in relatives of gastric carcinoma patients.15 Also, many patients with early gastric cancer, in particular of the diffuse type, do not have atrophy or intestinal metaplasia, and Hattori did not find precursor lesions in the margins of gastric microcarcinomas.16 Deterioration of corpus gastritis also facilitates the development of atrophy.11In a proportion of those infected, H pylorigastritis itself led to the development of atrophy and intestinal metaplasia, lesions that are recognised precursors of gastric cancer.17 The important question whether PPI treatment accelerates the development of atrophy cannot be answered definitively. The annual incidence of atrophy appears to be increased from 1–3% to 4–8% whenever a state of reduced acid secretion is present.18 In 1996, Kuipers et al published a report in which they compared two cohorts of GORD patients.19 One of these cohorts, in the Netherlands, received long term treatment with omeprazole while the other cohort, in Sweden, underwent antireflux surgery. During follow up, the annual incidence of atrophy was 6.1% of H pyloripositive patients receiving omeprazole while atrophy rarely developed in H pylori positive patients after surgery and in H pylori negative patients of both cohorts. The study has been criticised on methodological grounds. However, in other long term studies of PPI treatment, the incidence of atrophy in H pylori infected patients was consistent with Kuiper's study.11 ,20 In 1999, Lundellet al reported the eagerly awaited results of a randomised study comparing omeprazole maintenance therapy with antireflux surgery in patients with GORD.21 During the three year follow up, they did not find a significant excess of atrophy development in H pylori positive patients receiving omeprazole. However, this study too had several methodological flaws and reanalysing the data adds support to Kuipers' study rather than disproving it.21-25 At this point in time, we do not know whether eradicating H pylori infection before long term treatment with a PPI actually reduces the risk of atrophy development. We do however know from a randomised controlled study that cure of infection prevents deterioration of the corpus gastritis.26

In conclusion, treatment of H pyloriinfection can, in individual cases, lead to improvement in reflux symptoms although a clinically relevant effect on the natural history of GORD has not been proved with certainty. It is conceivable that in individual cases the dose of PPI needs to be adjusted following eradication of the organism. Simply ignoring the H pylori infection in patients with GORD and on long term PPI treatment inevitably leads to aggravation of corpus gastritis and probably also to acceleration of the development of atrophy. Both a more severe corpus gastritis and atrophy are indicators of an elevated risk of gastric cancer. Treatment with omeprazole can be used without risk for at least 10 years, in patients with H pylori infection also. However, whether this also applies to longer periods of treatment is as yet unclear. Presumably, a randomised controlled study investigating this long term safety aspect will never be carried out. As our current knowledge suggests that apart from the side effects of treatment no clinically relevant disadvantages are to be expected by eradicating infection in patients with GORD, the cautious physician will search for and eradicate H pylori, in particular in younger GORD patients requiring long term PPI treatment. Although this approach heals antral and corpus gastritis, this recommendation is not evidence based. It should however be pointed out that in other areas, chronic infections (for example, hepatitis C) associated with only a potential for fatal outcome are currently being treated with side effects associated expensive drugs despite a lack of evidence based proof of a long term beneficial effect of the treatment on the prognosis of the disease.