Discussion 2

Question : When should esomeprazole be given − in the morning before or after breakfast, or in the evening?

Prof. Dent : Esomeprazole should be given in the morning to target reflux during the daytime, which is the predominant period of reflux. There are data comparing the effects of esomeprazole on pH control when taken prior to and after food. There is no specific recommendation in the prescribing information, but I think that it makes sense to take the agent before breakfast, as it is easy to remember to take it then and there are sound theoretical, pharmacological and pharmacodynamic reasons for doing this.

Question : Should we worry about the presence of Helicobacter pylori in patients with reflux disease? Should we eradicate the bacterium before we treat these patients?

Prof. Vakil : I think that we need to separate population-based studies from studies performed in cohorts of patients. There is a negative association between H. pylori infection and GERD in populations. Most of the new data do not support the original view that eradication of H. pylori causes reflux oesophagitis. There are two ways that you can approach the issue of H. pylori in this setting. The first, which I think is followed more commonly in Europe, is that the H. pylori status is routinely assessed during endoscopy. H. pylori is a carcinogen and a pathogen, and is therefore routinely treated. In the USA, however, we follow a slightly different policy. When we perform endoscopy, we do not routinely test for H. pylori unless there is a lesion that gives us a reason to do so − for example, if a duodenal ulcer, erosive duodenitis or erosive gastritis are present. One can make a case for either of these approaches at the moment, we just do not know the answer.

Question : How should we manage Barrett's oesophagus in the long term?

Prof. Dent : The question of whether you should use a very high level of acid suppression in patients with Barrett's oesophagus will be discussed later. I do not think that the issue is settled at all. Undoubtedly, surveillance of these patients provides some degree of protection against the risk of developing adenocarcinoma. However, it is controversial as to whether this investment is worthwhile, and these issues will also be discussed later.

Question : We have heard today that the higher acid-suppressive dose of an agent is more efficient in the long run, but is this a good strategy? Should we not be using the least efficient dose for every person, so that as much acid as possible remains in the stomach, because acid is a good thing?

Prof. Lauritsen : There are many different management strategies. A ‘step-up’ approach is a definite possibility, but is unsatisfactory for the patient. The ‘step-down’ approach, starting with a powerful gastric acid-suppressive agent and then reducing the dose in order to reduce the cost of treatment, is preferred in most countries today, based on current management guidelines.

Prof. Dent : The key thing is to make a clear distinction between the strategies for initial therapy and those for long-term therapy. You can achieve the aims of initial therapy within 2–4 weeks in most patients using high-level acid suppression, and I do not think that anyone has concerns about this. You can then tailor treatment to the individual patient's needs, once you have achieved the initial therapeutic aim.

Question : I want to press Professor Vakil a bit more on the role of H. pylori in reflux disease, particularly in patients on long-term proton pump inhibitor therapy. There is some evidence, for example from Kuipers et al ., 1 that these patients may develop gastric atrophy, which could be a neoplastic condition. In these individuals, treating with a proton pump inhibitor could cause gastric cancer to develop in the future. I realize, however, that there are other data that are less conclusive, such as the Lundell et al . data. 2 What is your opinion on this?

Prof. Vakil : The question is if the patient is H. pylori- positive and you prescribe proton pump inhibitor therapy, are you going to accelerate the development of atrophic gastritis and/or increase the risk of gastric malignancy? With regard to gastric cancer, we can say that despite use of proton pump inhibitors for 10–15 years, we have not seen an increase in the incidence of gastric cancer in these patients. However, things might change after 30 years. With regard to the gastritis question, there are conflicting data. In the better-controlled study by Lundell et al . 2 proton pump inhibitor use did not increase the rate of gastric atrophy when patients were compared with a surgically treated group who did not receive proton pump inhibitors. One can argue for eradicating H. pylori in all cases to prevent gastric atrophy, but I do not think that a deliberate strategy to look for H. pylori infection in patients about to embark on proton pump inhibitor therapy is warranted at the present time. I do think that it is reasonable to discuss this with the patients and, if they want to be tested for H. pylori, one should test and treat them.

Prof. Talley :  H. pylori is starting to disappear in many parts of the Western world, so this is becoming a little less of an issue. We are seeing much less H. pylori now − I am almost starting to feel that testing is perhaps not worthwhile, and that is a big change for us.

Question : I would like to question Prof. Vakil even further on this issue. The recent Uemura trial, 3 published in the New England Journal of Medicine, looked at H. pylori in nonulcer dyspeptics. If the bacterium was not eradicated, some of these patients developed cancers. How do you justify to your patients, who do not have ulcers, that you are not going to test for H. pylori ? These patients are so alert and are able to obtain information, particularly from the Internet, and they will keep asking you why you are not testing for this bacterium.

Prof. Vakil : We would look proactively for patients who have a family history of gastric cancer and we would treat these patients. There is a European study of healthy H. pylori -positive blood donors, which showed that atrophy and intestinal metaplasia developed in up to 20% of cases over 10–15 years in Caucasians. This strengthens the argument for eradicating H. pylori routinely. However, these are only arguments, and we do not have clear data that this strategy is beneficial − it has recently been shown that atrophy can regress, but intestinal metaplasia does not do that. Do we actually change the natural history of the disease when we eradicate H. pylori in a 50-year-old patient when there are signs that intestinal metaplasia is already present? We do not know the answer.

Question : Are there any changes in laboratory parameters during long-term esomeprazole therapy?

Prof. Lauritsen : Some coincidental variations in laboratory parameters were seen during long-term therapy, but these were not considered clinically significant, as opposed to the major consistent finding of a modest increase in serum gastrin levels during short-term treatment. This was also seen during previous investigations of proton pump inhibitors, 4 and no doubt is probably related to the extent of gastric acid inhibition. There were no differences in gastritis scores or enterochromaffin-like-cell content, which could have been expected. No clinically significant increase occurred in the study of more than 2000 patients with reflux oesophagitis.

Question : In patients with unhealed oesophagitis, was there any disconnect between oesophagitis and symptoms?

Prof. Dent : Symptom relapse is a reasonably good predictor of relapse of oesophagitis after initial treatment success. However, the question is being asked in a slightly different way here, and I think that it would be interesting to hear from Professor Vakil on the results of the EAZEE study and what insights we have gained from these.

Prof. Vakil:  If a patient was free of symptoms, there was a predictive power of about 85% that the oesophagitis had healed. 5 It is not really surprising that there is a disconnect between nonhealing and symptoms. The classic studies by Baldi et al . 6 have shown that 85% of reflux episodes are not associated with symptoms, and this effect is particularly marked in patients with mild grades of oesophagitis. An even more interesting observation from the EAZEE trial is the fact that sustained resolution of heartburn predicts freedom from heartburn symptoms for the remainder of the trial with a high degree of accuracy. Ninety-five per cent of patients reaching this end-point were free of heartburn for the rest of the study. 7

Question : Would the panel like to comment on alternate-day maintenance therapy and weekend therapy? Are these strategies alternatives to daily maintenance therapy?

Prof. Lauritsen : Weekend omeprazole therapy was not very effective in patients with mucosal lesions, such as those with Los Angeles Grades A–D. 8 I think that alternate-day or weekend therapy could be an option in patients with endoscopy-negative reflux disease, which comprises a major proportion of patients in primary care. However, I do not think that these approaches are attractive when compared with on-demand therapy, with which the patient is the master of his/her treatment.