Dyspepsia is very common, and most patients who have persisting or intrusive symptoms will eventually require investigation. The most useful investigation to define disease in the upper gastrointestinal tract is endoscopy. Duodenal ulcers usually involve Helicobacter pylori, so treatment should aim to eradicate the organism and heal the ulcer. Gastric ulcers which are unrelated to non-steroidal anti-inflammatory drugs (NSAIDs) should be treated in the same way. NSAID-associated ulcers are best healed by stopping the drug, but if it is continued, healing can still occur if acid suppression is effective. Misoprostol provides significant protection against the development of gastric ulcer in patients taking NSAIDs. The cornerstone of treatment of moderate to severe gastro-oesophageal reflux disease is acid suppression with proton pump inhibitors. These need to be continued indefinitely at the healing dose to prevent relapse
Most adults experience dyspeptic symptoms from time to time. If these symptoms are sufficiently persistent or intrusive, and are not improved by over-the-counter remedies, medical advice will usually be sought. The issue confronting the family practitioner is whether to continue to treat the patient empirically (albeit more effectively with medications such as H2 receptor antagonists) or to investigate what disease process is going on. The timing of investigation is a matter of clinical judgement, but for most patients with ongoing or recurrent dyspeptic symptoms, it will eventually be necessary either to rule out serious disease or to allow a management plan to be prepared.
The problem with treating without any investigation is that dyspeptic symptoms are not as specific for particular diagnoses as some textbooks would lead us to believe. For example, gastroenterologists are well aware of how frequently heartburn is related to an ulcer, and epigastric discomfort to reflux disease. Moreover, serious disease can be present with minimal symptoms, particularly in older patients.
The most effective investigation is upper gastrointestinal endoscopy. Nowadays this is a safe and well tolerated procedure and is increasingly available 'on request'. Whether to make use of open access endoscopy or to seek, in addition, a specialist consultation is a matter for judgement in each individual case. Barium studies are seldom definitive and are probably most helpful in locations where endoscopy is not readily available.
Perhaps the main reason for having a firm diagnosis is that the approaches to management of the different conditions are now diverging. No longer is it appropriate simply to prescribe indefinite (intermittent or continuous) H2 antagonist therapy on the grounds that reasonable symptomatic relief is a fair enough outcome. For example, there is now the possibility of 'curing' ulcer disease, so the approach to treating a duodenal ulcer is becoming quite different from that of managing oesophageal reflux disease. Greater precision in diagnosis is therefore required to ensure optimal management, and endoscopy, for the most part, will provide that precision.
The treatment of duodenal ulcer in the mid 1990s is different from that of a decade ago. In 1985, Helicobacter pylori had only recently been discovered and there was minimal, if any, evidence to support the view that it was a cause of duodenal ulcer. Healing was primarily with H2 receptor antagonists, although site protecting drugs such as sucralfate or colloidal bismuth were also used. Even antacids were known to be effective healing agents.
Although most ulcers could be readily healed, there was a high relapse rate once treatment was stopped (up to 80% within 12 months). Ongoing management was usually watchful expectancy in the 20% of patients who did not relapse in 12 months, intermittent therapy based on symptoms in those who relapsed once or twice per year, and continuous maintenance treatment in those who relapsed promptly. Maintenance treatment with a half healing dose had a reduced, but still substantial, 12 month relapse rate (40%), while full healing dose maintenance reduced this further to about 20%.
There is now a broad consensus that a different approach is needed. H. pylori is now accepted as the cause of most duodenal ulcers. 'No H. pylori, no ulcer' holds true for all duodenal ulcers except those few related to nonsteroidal anti inflammatory drugs (NSAIDs) and the rare Zollinger-Ellison syndrome. The general rule should now be that, when patients with a duodenal ulcer are first diagnosed by endoscopy, H. pylori eradication should be attempted. Eradication (the absence of H. pylori from the antral mucosa one month after the cessation of eradication therapy) leads to very low relapse rates (2-3% per year). Relapses are always associated with renewed infection.
The standard H. pylori eradication regimen is triple therapy with bismuth subcitrate, tetracycline and metronidazole for two weeks (see box), although one week's treatment may be sufficient. This is effective in approximately 90% of patients. However, it is not particularly well tolerated, with 30% of patients complaining of adverse effects including nausea, diarrhoea, discoloured tongue and teeth. Amoxycillin substituted for tetracycline is not quite as efficacious (75-80% eradication).
Standard triple therapy (two weeks)
Other regimens recently investigated include a combination of a proton pump inhibitor (e.g. omeprazole or lansoprazole) and an antibiotic (e.g. amoxycillin or clarithromycin). This regimen is better tolerated, but efficacy is less predictable and somewhat lower (80-85% eradication at best). For best results, omeprazole should be given twice daily (20 mg or 40 mg twice daily) and it should be started with the antibiotic. Recent studies suggest that a combination of a proton pump inhibitor and two antibiotics, e.g. clarithromycin and metronidazole, may be even more effective (90-95%) than standard triple therapy, and that one week of treatment is sufficient.1 Healing the ulcer first with omeprazole and then adding the antibiotics is less successful. Omeprazole alone suppresses H. pylori without eradicating it, and appears to render the organism less susceptible to eradication by antibiotics. Eradication with H 2 receptor antagonists and two antibiotics is also effective (85-90%),2 but the period of treatment may need to be longer than with proton pump inhibitors.
There are many variations on the basic regimens which appear to be similarly effective, but none is completely safe or 100% successful. The optimal regimen has not yet been agreed on.
Should healing of the ulcer and eradication of the organism be confirmed in patients who have duodenal ulcers by further endoscopy and biopsy one month post-treatment? With patients who had an uncomplicated, symptomatic ulcer, it is probably reasonable to be guided by their symptoms. If they stay well after treatment, it can be assumed that H. pylori has been eradicated. Where there has been a complication such as a major bleed, particularly if the patient has not had ulcer pain, it would be wise to be more rigorous in ensuring that healing and H. pylori eradication have been achieved.
There is still, perhaps, a minor place for maintenance therapy to suppress acid secretion. This is particularly so for the older patient with multiple medical problems whom you judge to be unlikely to tolerate eradication therapy. Maintenance treatment should, however, be a full-healing, not half-healing, dose to minimise the likelihood of relapse.
The evidence linking H. pylori with benign gastric ulcer is less convincing than for duodenal ulcer. However, the consensus now is that the organism is probably important in the pathogenesis of the 70% of gastric ulcers not attributable to the use of NSAIDs. Eradication should therefore be attempted at first diagnosis, as for duodenal ulcer, using the same regimens. While duodenal ulcers can be assumed to be H. pylori related, for gastric ulcers the infection should be documented, given the less than one to one relationship. Healing of the ulcer and eradication of H. pylori should also be checked by endoscopy and biopsy one month after treatment has finished.
Ulcers due to NSAIDs heal most effectively if the NSAID is stopped. However, adequate healing rates can be achieved with acid suppressing drugs, particularly omeprazole, even when NSAIDs have to be continued. Remember that prophylactic H2 receptor antagonists do not provide protection against the development of gastric ulcer in patients taking NSAIDs. They do provide protection against NSAID-related duodenal ulcer, but this is a much less frequent complication. Misoprostol, a prostaglandin analogue, does provide some protection against the development of NSAID-related gastric ulcer. On average, 20% of patients on long-term NSAIDs will have a gastric ulcer. This can be reduced by half with the concomitant use of misoprostol (200 micrograms 4 times daily), and complications (bleeding and perforation) are reduced to a similar degree. Unfortunately, misoprostol prophylaxis may not be cost-effective in all patients.3 The efficacy of omeprazole in protecting against NSAID-related ulcers is presently under study.
Maintenance acid suppression for ongoing treatment of gastric ulcer probably still has a limited place. The indications would be similar to those for duodenal ulcer.
Gastro-oesophageal reflux disease
In contrast to ulcer disease, efficient acid suppression is the key to the medium and longer-term management of the more severe grades of reflux oesophagitis. Many people experience reflux symptoms, but they are usually not particularly severe or frequent. Endoscopy to assess oesophageal pathology should be considered if
- symptoms are marked or longstanding or atypical
- the patient is elderly
- there has been any suggestion of gastrointestinal bleeding
- there has been any difficulty swallowing.
Unpleasant reflux symptoms can occur without any abnormal endoscopic findings. In this case, 24-hour ambulatory pH monitoring may help to identify any reflux.
There are a number of simple measures which may help to relieve minor reflux symptoms.4 These include stopping smoking, taking alcohol and coffee in moderation, avoiding stooping, avoiding large meals (particularly at night), avoiding foods which invariably precipitate symptoms, sleeping propped up and losing weight.
Many patients resort to antacids which usually produce rapid, if transient, symptomatic relief. For mild to moderate reflux disease, H2 receptor antagonists are usually effective. They should be given twice daily for best effect. Double doses, e.g. ranitidine 300 mg twice daily, may give an extra measure of healing and symptomatic relief when normal doses have been only partially effective. However, H2 receptor antagonists are generally disappointing from both the symptomatic and healing points of view in more severe reflux disease.
Drugs which assist the motor function of the stomach or oesophagus, such as metoclopramide and cisapride, can also be effective in treating reflux oesophagitis. However, they are no more effective than H2 receptor antagonists and, even in combination with H2 receptor antagonists, they do not appear to deal with more severe reflux disease.
The acid pump inhibitors such as omeprazole or lansoprazole have revolutionised the treatment of more severe grades of reflux disease. This is due to the more effective and prolonged (>24 hours) suppression of acid secretion. Omeprazole 20 mg or lansoprazole 30 mg in the morning will heal oesophagitis in the majority of patients and provide prompt symptomatic relief. Some patients will need a larger daily dose. Even patients with strictures who had previously required frequent dilatations will usually become asymptomatic and need no more dilatations.
However, indefinite maintenance treatment using the full healing dose is required to prevent relapse. This raises the issue of the role of surgery in the younger patient. The prospect of a lifetime - several decades - of continuous acid suppression may be a less attractive option than definitive surgery. A Nissen fundoplication, increasingly undertaken laparoscopically, is a reasonable alternative to prolonged acid suppression. There is a small operative mortality and patients need to be warned that they may be unable to belch or vomit after the operation. Most patients in older age groups will probably be best treated with long-term omeprazole; the younger patient might be better served by surgery.
Jones R. Dyspeptic symptoms in the community. Gut 1989;30:893-8.
Brown C, Rees WD. Dyspepsia in general practice [see comments]. Br Med J 1990;300:829-30. Comments in: Br Med J 1990;300:1137, 1340.
Forbes G. Helicobacter pylori eradication: who, why and how in 1994? Med J Aust 1994;161:291-2.
Elliott SL, Yeomans ND, Buchanan RR, Smallwood RA. Efficacy of 12 months' misoprostol as prophylaxis against NSAID-induced gastric ulcers. A placebo-controlled trial. Scand J Rheumatol 1994;23:171-6.
Hetzel DJ, Dent J, Reed WD, et al. Healing and relapse of severe peptic oesophagitis after treatment with omeprazole. Gastroenterology 1988;95:903-12.
The following statements are either true or false.
1. The combination of an antibiotic and a proton pump inhibitor is as effective as a combination of bismuth subcitrate, tetracycline and metronidazole for the eradication of H. pylori.
2. Most gastric ulcers are due to NSAIDs.
Answers to self-test questions
- Labenz J, Stolte M, Ruhl GH, Becker T, Tillenburg B, Sollbohmer M, et al. One-week, low-dose triple therapy for the eradication of Helicobacter pylori infection [see comments]. Eur J Gastroenterol Hepatol 1995;7:9-11. Comment in: Eur J Gastroenterol Hepatol 1995;7:1-3.
- Hentschel E, Brandstatter G, Dragosics B, Hirschl AM, Nemec H, Schutze K, et al. Effect of ranitidine and amoxycillin plus metronidazole on the eradication of Helicobacter pylori and the recurrence of duodenal ulcer [see comments]. N Engl J Med 1993;328:308-12. Comment in: N Engl J Med 1993;328:349-50. Comments in: N Engl J Med 1993;329:59-60, 1356.
- Day R, Henry D. Do anti-ulcer drugs prevent gastrointestinal damage from NSAIDs? [editorial]. Aust Prescr 1991;14:42-3.
- Yeomans ND. Gastro-oesophageal reflux disease. Aust Prescr 1992;15:34-5.