Treatment of Ulcers Associated with NSAIDs


Synopsis: Acid-suppressing drugs are effective in healing NSAID-related ulcers and erosions, in relieving NSAID-related symptoms, and in preventing recurrence of NSAID-related pathology following initial healing when these drugs are continued.

Sources: Hawkey CJ, et al. N Engl J Med 1998;338:727-734; Yeomans ND, et al. N Engl J Med 1998;338:719-726.

Two related studies (termed the omnium and ASTRONAUT studies) examined the effects of a variety of regimes on ulcers related to nonsteroidal anti-inflammatory drugs (NSAIDs). The first study was a double-blind comparison of 935 patients who required continuous NSAID therapy and who had ulcers or 10 or more erosions in the stomach, duodenum, or both. The subjects were treated with omeprazole 20 or 40 mg daily or misoprostol 200 mcg qid. End points were healing of ulcers or erosions and relief of symptoms. Those healed at four or eight weeks were then randomized to maintenance therapy with omeprazole 20 mg daily, misoprostol 200 mcg daily, or placebo and evaluated at six months. They found that the overall rates of successful treatment of ulcers, erosions, and symptoms associated with NSAIDs were similar for the two doses of omeprazole and for misoprostol. Therapy with omeprazole was associated with a lower rate of relapse than misoprostol (% in remission at 6 months, 61% vs 48%). Omeprazole was better tolerated than misoprostol (% with adverse events during the healing phase, misoprostol 59%, omeprazole 48% and 46%).

The second study involved 541 patients who required continuous treatment with NSAIDs and who had ulcers or more than 10 erosions in either the stomach or duodenum. Patients were randomly assigned to double-blind treatment with omeprazole 20 mg or 40 mg orally per day or ranitidine 150 mg orally bid for 4-8 weeks. Response, defined as "healing of ulcers and erosions and improvement in symptoms," was assessed at four and eight weeks. Those who responded were then randomized to maintenance treatment with either omeprazole 20 mg daily or ranitidine 100 mg bid and evaluated at six months. The initial treatment response was significantly better with both doses of omeprazole than ranitidine; treatment was successful in 80% of those on omeprazole 20 mg daily, 79% of those given omeprazole 40 mg daily, and 63% of those given ranitidine. The relapse rate at six months in the omeprazole group was 72% and, in the ranitidine group, 59%. Both medications were equally well tolerated.


NSAID-related gastrointestinal symptoms and complications represent a significant health problem. Chronic NSAID use is indicated for a large number of patients with chronic inflammatory arthropathies, and it is clear that this use is associated with significant gastrointestinal symptomatology. Of considerable concern is the increased prevalence of gastrointestinal hemorrhage and upper gastrointestinal perforation in relation to chronic NSAID use. The risk for, and mortality related to, these complications is particularly significant among the elderly, in whom chronic NSAID use is a major contributor to gastrointestinal hemorrhage and perforation. While attempts have been made to modify NSAID formulation and to minimize these risk factors, they continue to represent a major problem. Given the role of prostaglandin metabolism in the pathogenesis of NSAID-related GI side effects, considerable effort was placed in the development of prostaglandin analogs in an attempt to reduce or eliminate these side effects. Prospective studies suggested efficacy for misoprostol, but its use has been complicated by side effects and expense. Previous studies, as well as anecdotal experience, have suggested that acid-suppressing agents, such as H2 receptor antagonists and proton pump inhibitors, might also be effective.

These large, well-conducted studies provide several answers of clinical importance. First, acid-suppressing drugs are effective in healing NSAID-related ulcers and erosions, in relieving NSAID-related symptoms, and in preventing recurrence of NSAID-related pathology following initial healing when these drugs are continued. Second, the proton pump inhibitor, omeprazole, in a dose of 20-40 mg daily, is more effective than ranitidine, the H2 receptor antagonist, in healing NSAID-related ulcers and erosions. It is also more effective in the long-term prevention of relapse of these lesions. It is, therefore, clearly to be preferred over the H2 receptor antagonist in this area. Third, though the initial healing rates for misoprostol and omeprazole were similar, omeprazole was more effective in the prevention of relapse of NSAID-related lesions and was better tolerated. For these reasons, I believe that omeprazole now becomes the agent of choice for the treatment of, and prevention of, NSAID-related upper gastrointestinal ulcers and erosions. We should be aware, however, that this study does not provide all of the answers. Thus, we do not know if any of these approaches would be effective in reducing the prevalence of gastrointestinal hemorrhage or perforation related to ulcers or erosions. One would predict that it should, but clearly this needs to be demonstrated in a prospective study. Whether this will be ever achieved, given the large number of patients that would need to be studied to show this, is uncertain. In any event, I believe that these studies now provide clear guidelines for the management of those patients who take NSAIDs and who will continue to require their use in the long term.


    1. Dajani EZ. Prostaglandins, NSAIDs and GI mucosal integrity: Can we identify patients at risk of NSAID-induced gastric injury? Am J Gastroenterol 1996;91:835-836.

    2. Hirschowitz BI. Non-steroidal antiinflammatory drugs and the gastrointestinal tract. Gastroenterol 1994;2:207-223.