Article Text
Statistics from Altmetric.com
See article on page 473
Functional dyspepsia is a common disorder of heterogeneous aetiology for which there is no clearly effective treatment. In this issue of Gut, Blum and colleagues1 report their large controlled clinical trial assessing the effectiveness of antisecretory medication in functional dyspepsia (see page 473). Patients with and without evidence ofHelicobacter pylori infection were recruited into the study.
The study found symptomatic benefit over placebo at the end of two weeks' treatment with omeprazole 20 mg per day inH pylori positive patients but no significant benefit in primary outcome with the same treatment inH pylori negative patients. The therapeutic gain over placebo for omeprazole 20 mg in the H pylori positive patients was 17.6% (95% confidence intervals 4.2–31%; p<0.014). There was no significant benefit over placebo following omeprazole 10 mg or ranitidine 150 mg inH pylori positive patients or following any of the three treatments in the H pylorinegative subjects. The superior benefit from omeprazole 20 mg in theH pylori positive versus negative subjects was also apparent with respect to improvement in quality of life.
The reason for symptomatic benefit over placebo with powerful acid suppression being confined to the H pyloriinfected subjects in this study is unclear and needs to be addressed. An earlier large study in which patients with functional dyspepsia were treated with omeprazole 20 mg/day found a 10% benefit over placebo but no relationship to H pylori status.2
Several aspects of the design of the study by Blum and colleagues may have contributed to the different responses seen in theH pylori positive and negative subjects. The investigators were not blinded to the H pylori status of the patients. In addition,H pylori positive patients not responding to treatment were eligible for recruitment into a second study examining the effect of H pylori eradication therapy; an option not open to H pylori negative non-responders. The desire to recruit patients into the second study (and any rewards for doing so) may have biased the investigators in favour of categorising the H pylori positive subjects as non-responders. The markedly lower response to placebo, as well as to all other forms of treatment in the H pylori positive subjects compared with negative subjects, suggests that investigators were indeed much less inclined to recognise a beneficial response in the infected group. The placebo response inH pylori positive patients was only 42% compared with 66% in the H pylori negative patients. The magnitude of the reduced response to placebo in theH pylori positive subjects was actually greater than the therapeutic benefit of omeprazole over placebo in that group. This led to the anomalous situation where the proportion ofH pylori positive subjects showing a beneficial response following omeprazole treatment (59%) was substantially less than that of H pylorinegative subjects showing a beneficial response after omeprazole (71%) despite the effectiveness of omeprazole over placebo being confined to the former group.
The much lower placebo responses recorded in theH pylori positive subjects may have contributed to the benefit of omeprazole over placebo being apparent only in the H pylori positive subjects. The ability to detect the benefit of any active treatment is reduced when there is a large placebo response as it can only be observed in the proportion not responding to placebo. Blinding the investigators toH pylori status and removing any incentive for categorising H pylori positive subjects as non-responders may have produced a different result.
Other reasons for benefit over placebo being apparent only in theH pylori positive subjects need to be considered. The authors propose that it was due to the more profound elevation of intragastric pH induced by omeprazole inH pylori positive versus negative subjects. However, the effect of this difference in pH control with respect to symptom control and mucosal healing is small. A recent large study by Hatlebakk et al in patients with heartburn indicated that omeprazole 20 mg/day produced resolution of symptoms in 86% of H pylori positive subjects versus 65% of H pylorinegative.3 Similarly, Holtmann et al observed that relief of heartburn with pantoprazole in patients with oesophagitis was achieved in 89.2% ofH pylori positives versus 84.5% ofH pylori negatives.4 This small incremental benefit could not explain the efficacy of omeprazole over placebo being confined to H pylori positive subjects in the present study.
Blum et al propose that theH pylori infected mucosa may result in enhanced sensitivity to gastric acid and that this may explain the benefit of omeprazole being confined to H pylori infected subjects. Such a condition may be a precursor to peptic ulceration. It is well recognised that a proportion ofH pylori positive patients with functional dyspepsia develop frank peptic ulcer disease in the near future5-7 and the proportion doing so is similar to the proportion of H pylori positive subjects who responded to omeprazole in this study. The subgroup ofH pylori positive functional dyspepsia patients with this pre-ulcer condition may account for the beneficial response to H pylori eradication in some subjects with functional dyspepsia.5 ,6 If the benefit of omeprazole in the H pylori positive subjects is due to either an acid sensitive mucosa and/or a pre-ulcer state, then an equivalent and more permanent beneficial response could be achieved in the H pylori positive subjects by eradicating the infection and restoring a normal mucosa.
Concentrating on the different effects of omeprazole inH pylori positive and negative subjects should not distract from another important finding in this study. The study provides useful information on the relative efficacy of different treatments for patients presenting with functional dyspepsia. It demonstrates that placebo plus simple antacid therapy is a highly effective form of treatment for functional dyspepsia with a response rate of 42% in H pylori positives and 66% in H pylori negatives. The next most effective factor influencing response may be the doctor's desire for a particular outcome as this may well have accounted for the 20% superior response to all forms of treatment in theH pylori negative versus positive subjects. Both of these responses are consistent with functional dyspepsia being a disorder with a substantial supratentorial component.
The main conclusion of this paper, that a relatively small proportion of H pylori positive but not negative patients with functional dyspepsia respond to omeprazole, seems robust. However, it is possible that this same beneficial response could be achieved and in a more permanent form by eradicating theH pylori infection.