Whom should we “test and treat” for Helicobacter pylori?

Peptic ulcer disease

H pylori is causally implicated in the pathogenesis of pyloric ulcer disease.1 A systematic review and meta-analysis of randomised controlled trials of eradication therapy versus placebo in H pylori positive disease showed that eradication therapy was cost effective compared with long term acid suppression and led to significantly lower rates of duodenal and gastric ulcer relapse, with numbers needed to treat (NNT) to prevent one ulcer relapse of 2 and 3 respectively.2

Gastro-oesophageal reflux disease

A systematic review of observational studies showed that the prevalence of H pylori was significantly lower in people with gastro-oesphageal reflux disease than in healthy controls, suggesting that infection may be protective.3 This is biologically plausible because H pylori can induce an atrophic gastritis, which may reduce production of gastric acid. However, the association could also be the result of a confounding by socioeconomic status (reflux disease is more common in higher socioeconomic groups,4 whereas H pylori infection is inversely associated with socioeconomic status). Two systematic reviews of randomised controlled trials and cohort studies have suggested that eradication therapy has no benefit in H pylori positive patients with reflux disease followed up for at least six months,5 6 although a third review disputed this finding.7 This evidence persuaded the Maastricht consensus group that H pylori infection is unlikely to have a major role in the disease,8 and we suggest that H pylori “test and treat” should not be used in patients whose predominant symptoms are heartburn as these patients are more likely to have gastro-oesophageal reflux disease.

Uninvestigated dyspepsia

Observational studies suggest that 5% of dyspepsia in the community may be attributable to H pylori.9 Treatment of H pylori infection should improve the symptoms of dyspepsia, by a combination of the healing of undiagnosed peptic ulcer and a small improvement in the symptoms of functional dyspepsia. A strategy of H pylori “test and treat” was therefore proposed for the initial management of uninvestigated dyspepsia, rather than prompt upper gastrointestinal endoscopy. Several large, pragmatic, randomised controlled trials have compared the two approaches. Meta-analysis of individual patient data from these trials showed that although prompt endoscopy led to a small but significant improvement in symptoms compared with test and treat, it was not cost effective for the initial management of dyspepsia, and “test and treat” is the preferred strategy.10

Functional dyspepsia

Three quarters of people with upper gastrointestinal symptoms who have upper gastrointestinal endoscopy will have no structural organic cause detected for their symptoms.11 If the main symptom in such patients is heartburn, they are said to have non-erosive reflux disease; in those with epigastric pain the diagnosis is functional dyspepsia.11 Numerous randomised trials have compared eradication therapy with placebo for H pylori positive functional dyspepsia, but results have been conflicting. However, a meta-analysis of these trials showed a significant benefit from treatment, with an NNT to cure one case of dyspepsia of 13 (95% confidence interval 9 to 19) compared with placebo.12 The effect on functional dyspepsia is modest, but short term eradication therapy can have a long term effect on the symptoms of dyspepsia so it is likely to be cost effective compared with treating with acid suppression.13

General population

Gastric neoplasia is the second commonest cause of cancer mortality worldwide, with almost 1 million cases and 0.75 million deaths annually.14 Systematic reviews of observational studies have reported a strong association between H pylori infection and gastric cancer,15 and the World Health Organization classifies H pylori as a human carcinogen.16 The key question, however, is whether eradication of the infection through population screening and treatment programmes reduces the incidence of gastric cancer. A systematic review of six randomised trials evaluating H pylori eradication compared with placebo in over 6000 people found a significant reduction in the incidence of gastric cancer.17 However, the finding is dependent on the results of one trial,18 and more data are needed before screening can be recommended in general populations. Two UK randomised trials suggested that a population H pylori screening and treatment programme reduces dyspepsia in those who receive eradication therapy.19 20 Longer follow-up in these trials has also suggested this could save money because less is spent on treating dyspepsia, and that this could partly offset any costs of a screening programme.21 22 More data are needed on the benefits and risks of population screening and treatment.

Harms of eradication

H pylori infection has been associated with a decreased risk of oesophageal adenocarcinoma (relative risk of oesophageal adenocarcinoma in H pylori positive individuals = 0.56; 95% confidence interval 0.46 to 0.68).23 Asthma has also been shown to have an inverse association with H pylori, although this correlation is weak.24 Epidemiological studies have also suggested that H pylori may increase the risk of a variety of diseases, from ischaemic heart disease to colon cancer.25 Many of these associations will be the result of confounding factors, but further research is important into diseases that could be influenced by H pylori infection.26

H pylori eradication therapy is clearly associated with an increased risk of adverse advents associated with antibiotics, including a risk of Clostridium difficile associated diarrhoea.27 Wide use of H pylori eradication is also likely to increase bacterial resistance to antibiotics in the general population.