Skip navigation

Helicobacter and dyspepsia

Study 1 [1]
Study 2 [2]
Results
Comment
Most people have dyspepsia at some time or another, and in the USA between 2% and 5% of all GP visits are for dyspepsia. When Helicobacter pylori infection of the stomach became established as a prime cause of peptic ulcers, it was hypothesised that eradication of the bacterium would not only cure ulcers, but also be beneficial for many people with dyspepsia. The problem was that most trials examining this question were flawed. Most did not even use an effective eradication regimen.

Two excellent studies [1,2] have now been published which compare the effectiveness of Helicobacter eradication with a short course of proton pump inhibitor, and which look at long-term (12-month) outcomes after eradication therapy in nonulcer dyspepsia. Both were randomised, double-blind and double-dummy. Both followed intensive clinical and endoscopic assessment of patients including multiple tests for Helicobacter pylori, clinical evaluations, and initial and follow-up endoscopy. They come to (apparently) different conclusions.

Study 1 [1]


This MRC-sponsored study in Glasgow enrolled 318 patients to receive omeprazole plus antibiotics or omeprazole alone for two weeks. The main endpoint was resolution of symptoms 12 months after treatment.

Study 2 [2]


The second study was jointly funded by The Swiss National Foundation and Astra, was multi-centre, and enrolled 328 patients to receive omeprazole plus antibiotics or omeprazole alone for one week. The main endpoint was relief of dyspeptic symptoms at 12 months after treatment.

Results


Eradication of Helicobacter pylori was about 80% or more with eradication therapy, and low with omeprazole alone.

The main outcomes for each trial are shown in the Table, together with the overall results of both trials combined. One trial [1] showed a statistical improvement for symptoms at 12 months. The other [2] did not. Overall, ten patients had to be given a short course of omeprazole plus antibiotics for one to have symptomatic relief from dyspepsia at 12 months who would not have had symptomatic relief with omeprazole alone.


Outcome McColl et al [1] Blum et al [2] Combined
Success with omeprazole (%) 7 21 14
Success with antibiotics & omeprazole (%) 21 27 24
NNT for one-year success 7.3 (4.7 to 17) 15 (6.3 to -40) 9.9 (6.2 to 25)
NNT for one-year healed gastritis   1.4 (1.3 to 1.5) 1.4 (1.3 to 1.5)

For healing of gastritis, the NNT with omeprazole plus antibiotics was 1.4 (95% confidence interval 1.3 to 1.5).

Comment


Why should one trial be positive and the other not? It may be that the populations being treated had different conditions, as an accompanying editorial suggests [3]. It may also be random chance. Both trials, and the combined result, show a moderate effect, mirrored in one [1] by reduced prescribing of antisecretory medicines. Then there is the way in which symptoms were scored. One study [2] used a score of 0 or 1 out of a scale of 0 to 7, while the other [1] used a score of 0 or 1 out of a scale of 0 to 20.

Is there enough here to make a judgement about treatment guidelines for dyspepsia? The answer is that on this evidence there is probably not. We would need a systematic review which would include at least one other high quality trial which has been done (and is reportedly positive), plus an economic evaluation, plus some supporting evidence to make a judgement.

The economic analysis would need to balance need for endoscopy (a significant proportion of patients would be over 45 years, when endoscopy is recommended) against lower costs of prescribed antisecretory medicines. The human analysis might take into account that eradication treatment is likely to prevent development of some peptic ulcers. But for now it seems to be 'watch this space'.

References:

  1. K McColl et al. Symptomatic benefit from eradicating Helicobacter pylori infection in patients with nonulcer dyspepsia. New England Journal of Medicine 1998 339: 1869-1874.
  2. AL Blum et al. Lack of effect of treating Helicobacter pylori infection in patients with nonulcer dyspepsia. New England Journal of Medicine 1998 339: 1875-1881.
  3. LS Friedman. Helicobacter pylori and nonulcer dyspepsia. New England Journal of Medicine 1998 339: 1928-1930.
previous or next story in this issue