Management of dyspepsia should be individualised

Editor—The papers by Chiba et al and McColl et al show that in the medium term a test and treat approach in dyspepsia can be as effective or better than endoscopically led management.¹,² However, an individualised approach remains prudent in primary care.

In the long version of the paper by Chiba et al on bmj.com (bmj.com/cgi/content/full/324/7344/1012), 33% of people with a positive result to the initial near patient test were negative on breath testing. Such results are important in populations where the prevalence of Helicobacter pylori is relatively low. Young adults with dyspepsia have a low pre-test probability of being H pylori positive in most developed countries and this, as McColl et al state, alters the investigation strategies used. In such patients a positive result for H pylori might well be a false positive. If the prevalence of H pylori is 25% then 11-12% of positive breath tests will be false positives, the false positive rate is 25% when serological tests are used, and with near patient tests it approaches 50%.³ Eradicating non-existent H pylori is unlikely to improve dyspepsia.

Secondly, the estimated 15% of people with undifferentiated dyspepsia who actually have a peptic ulcer will benefit most from a test and treat policy. The 25% whose symptoms are related to gastro-oesophageal reflux disease are unlikely to benefit. This leaves about 60% of people who have non-ulcer dyspepsia.

If we accept the results of a Cochrane review, the number needed to treat for H pylori eradication in non-ulcer dyspepsia is 15.⁴ Correspondence raised a number of concerns with the approach taken to produce this systematic review and its main result. In addition observational data link H pylori with gastric cancer but also indicate that a lack of H pylori is associated with gastro-oesophageal reflux disease, Barrett's oesophagus, and adenocarcinoma of the oesophagus and gastric cardia.⁵ A potential protective effect of H pylori against oesophageal cancer should be considered when making test and treat decisions when at best it appears that only one person out of 15 benefits symptomatically from eradication of H pylori in non-ulcer dyspepsia. H pylori will be successfully eradicated in most of the other 14, but with unknown long term consequences.

A test and treat approach can therefore be useful in the medium term for the management of dyspepsia for some people, but alternative strategies should be discussed with patients and an informed decision made on an individual basis.