Article Text


Helicobacter pylori and dyspepsia: trick or treat?
  1. Department of General Practice and Primary Care
  2. Guy’s, King’s and St Thomas’s School of Medicine
  3. 5 Lambeth Walk
  4. London SE11 6SP, UK

Statistics from

See article on page 186

Clinicians working in both primary and secondary care still do not have evidence-based guidance on the management of dyspepsia which is both clinically effective and cost-efficient. Perhaps we never will. Moving therapeutic targets often generate questions faster than researchers can answer them. Nevertheless, we do need better information about the inter-relations between dyspepsia,Helicobacter pylori infection,H pylori eradication, and endoscopy. The paper by Heaney et al (see page 186) provides a good deal of information, but also raises further questions about alternative strategies for managing dyspepsia.

The background is familiar. Until H pylori burst on to the scene, there was some controversy about empirical therapy versus early endoscopy in the initial management of dyspepsia, but the availability of accurate diagnostic and effective therapeutic methods in H pylori management means that clinicians can now choose from two further strategies—(1) test and endoscope and (2) test and treat. In the first of these theH pylori test is used as a “screen” for endoscopy, with H pylori positive patients undergoing endoscopy and H pylori negative patients treated as if they have non-ulcer dyspepsia. Patel and colleagues1 claimed that this approach would reduce the need for endoscopy by about 40%. The test and treat strategy, conversely, implies that uninvestigated dyspeptic patients (without alarm symptoms) should be tested for H pylori infection using the most accurate method available and those testing positive should undergo H pylori eradication therapy whereas those testing negative should be diagnosed as non-ulcer dyspepsia (and, presumably, gastro-oesophageal reflux disease as well).

Most of the information that we have had up until now, as Heaney and colleagues point out, has come not from prospective randomised trials, but from computer simulations of various kinds.2 3 Even this shadowy area of research has generated sparky controversy, with some groups concluding that there was little to choose in clinical or cost terms between the various treatments and others claiming therapeutic gain or cost savings, particularly from the test and treat approach.4 Heaney’s group are probably right in claiming that this is the first prospective randomised trial comparing a test and treat strategy with endoscopic “business as usual”.

One of the attractions of Heaney et al’s paper is that it dips its toe fairly determinedly in the potentially murky waters of health services research. As well as measuring conventional parameters including endoscopic findings,H pylori breath test results and symptom scores, the authors also used the SF36 to examine changes in quality of life and the Crown Crisp index to examine personality traits, although the latter does not seem to have generated any interesting information. Patients were followed for 12 months, which was a welcome extension of the normal length of follow up in H pylorieradication trials, but the study design could have been further strengthened by the incorporation of a formal health economic evaluation.

The results make interesting and stimulating reading. They are fairly persuasive that, taken in isolation from the possible disbenefits of eradication of H pylori and the over-use of antibiotics and community antibiotic resistance, a test and treat strategy is clinically appropriate and resource-sparing. Their conclusions are bolstered by the finding, presumably a surprise to the authors, of an excess of symptomatic improvement in the test and treat group over the endoscopy group. This observation is, incidentally, at odds with the received wisdom that a normal upper gastrointestinal endoscopy leads to reductions in anxiety, severity of symptoms and health care resource use.5 6 Other unexpected side effects of this paper are some useful information about responses to triple therapy given at various stages in the management ofH pylori positive patients and also the sensitivity to change of the SF36 in this group of patients without severe illness.

The authors provide an important caveat, which is that these results were obtained at a secondary referral centre and should not be extrapolated to primary care. They recommend the evaluation of a similar strategy in this setting. The results of a comparable controlled trial in primary care, which included a health economic evaluation, are currently in press, and broadly support the conclusions of Heaney’s group, adding to existing evidence of clinical effectiveness information about the cost-savings associated with empirical H pylori eradication in the uninvestigated dyspeptic patient.7

However, whether or not these findings should be translated into clinical practice depends on the answers to a number of research questions. Firstly, will we do a disservice to patients without duodenal ulcer disease when we eradicate their H pylori? Uncertainties still exist about the value ofH pylori eradication in non-ulcer dyspepsia,8 particularly in primary care patients in whom little research has been done. There are also concerns about the long term risks of gastro-oesophageal reflux disease and cancer afterH pylori eradication.9Prospective randomised studies in primary care with adequate length of follow up are required to settle these points. Secondly, does the “unnecessary” use of triple therapy (i.e. in non-ulcerH pylori positive patients) lead to unacceptable consequences to individuals (in terms of side effects of therapy) or to the community (in terms of increasing levels of antibiotic resistance)? Compared with the widespread use of these drugs in primary care to treat respiratory, gastrointestinal and pelvic symptoms and infections, the contribution of triple therapy is likely to be minimal, although clear information is still needed on this point. Thirdly, will the increasingly widespread use of a test and treat approach, particularly when it is based on suboptimal methods of detecting H pylori infection (such as some currently available near patient serology kits) lead to a less critical and potentially hazardous evaluation of dyspeptic patients? Perhaps, only time will tell, but H pylori testing must not be allowed to take the place of careful clinical assessment.

See article on page 186


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles