Current issues of ACP Journal Club are published in Annals of Internal Medicine


Omeprazole healed ulcers and maintained remission better than ranitidine

ACP J Club. 1998 Jul-Aug;129:9. doi:10.7326/ACPJC-1998-129-1-009

Source Citation

Yeomans ND, Tulassay Z, Juhász L, et al., for the Acid Suppression Trial: Ranitidine versus Omeprazole for NSAID-Associated Ulcer Treatment (ASTRONAUT) Study Group. A comparison of omeprazole with ranitidine for ulcers associated with nonsteroidal antiinflammatory drugs. N Engl J Med. 1998 Mar 12;338:719-26.



Are ranitidine and omeprazole comparable for healing and maintaining remission in patients with gastroduodenal ulcers and erosions associated with nonsteroidal anti-inflammatory drugs (NSAIDs)?


Randomized double-blind trial.


73 centers in 15 countries.


541 patients aged 18 to 85 years (mean age 57 y, 67% women) who required NSAIDs at standard therapeutic doses and not > 10 mg/d of prednisolone or its equivalent and had ulcers ≥ 3 mm in diameter or > 10 gastric or duodenal erosions. Exclusion criteria were contraindications to endoscopy, pyloric stenosis, erosive or ulcerative esophagitis, major active gastrointestinal hemorrhage, or disorders that could modify absorption of study drugs.


In the treatment phase, 174 patients were allocated to omeprazole, 20 mg/d; 187 to omeprazole, 40 mg/d; and 174 to ranitidine, 150 mg twice/d, for 4 weeks. This was extended to 8 weeks if healing did not occur. In the maintenance phase, 432 patients (398 who were considered healed) were enrolled. 210 were allocated to omeprazole, 20 mg/d, and 215 to ranitidine, 150 mg twice/d. Maintenance was done for 6 months.

Main outcome measures

Predefined treatment success at 8 weeks and maintenance of remission at 6 months.

Main results

Efficacy analysis included 535 patients in the treatment phase and 425 in the maintenance phase. Healing rates at 8 weeks were 80% and 79% in the low- and high-dose omeprazole groups, respectively, and 63% in the ranitidine group (P ≤ 0.001 for comparisons with ranitidine) (Table). At 6 months, patients in the omeprazole group stayed in remission at a higher rate than did patients in the ranitidine group (P = 0.004) (Table). Adverse event rates were high in both phases, but the rates did not differ significantly among groups (30% and 38% in the low- and high-dose omeprazole group and 40% in the ranitidine group for the treatment phase; 64% in the omeprazole group and 58% in the ranitidine group for the maintenance phase).


In patients who used high-dose NSAIDs regularly for 6 months, omeprazole was more effective than ranitidine at healing ulcers and maintaining remission. Adverse event rates were high but similar in all groups.

Source of funding: Astra Hässle.

For correspondence: Dr. N.D. Yeomans, Department of Medicine, University of Melbourne, Western Hospital, Footscray, Victoria 3011, Australia. FAX 61-3-9318-1157.

Table. Omeprazole (OM) vs ranitidine for healing and maintaining remission in nonsteroidal anti-inflammatory drug-induced ulcers*

Comparison Purpose OM Ranitidine RBI (95% CI) NNT (CI)
Low-dose OM vs ranitidine Healing 80% 63% 27% (12 to 46) 6 (4 to 13)
High-dose OM vs ranitidine Healing 79% 63% 25% (10 to 44) 6 (4 to 15)
OM vs ranitidine Remission 72% 59% 22% (6 to 40) 8 (5 to 26)

*Abbreviations defined in Glossary.


Omeprazole maintained remission from ulcers better than misoprostol

Despite many studies over the past 3 decades that repeatedly highlighted the gastroduodenal side effects of NSAIDs and their serious complications (hemorrhage, perforation, and death), no evidence exists to suggest a decrease in their use or side effects. In the absence of a breakthrough in the treatment of arthritic diseases, NSAIDs are likely to stay in common use well into the next century. Management of NSAID-related ulcers and erosions remains a clinical challenge and should include prevention (prophylaxis), healing, and maintenance of remission.

Misoprostol and ranitidine have dominated this field over the past 10 years. Their use has been the subject of debate among researchers and clinicians alike. Misoprostol is inexpensive and prevents gastric and duodenal ulcers and their complications, but its use can be limited by the abortifacent activity in young women. It is contraindicated in the absence of contraception in these women and by the development of transient abdominal discomfort and diarrhea, which occurs in up to 30% of persons in most clinical studies. Ranitidine is better tolerated but at normal doses only prevents duodenal ulcers—NSAID ulcers are more likely to affect the stomach.

While the debate on ulcer prophylaxis continues, a more pressing clinical question is how to heal an ulcer that has already developed. The place of misoprostol and ranitidine in healing NSAID-induced peptic ulcers is not well established, and healing can be slowed by continued NSAID intake. New approaches are needed, and therefore the studies by Hawkey and Yeomans and their colleagues are important. Patients in these international multicenter studies were carefully selected through use of similar and well-defined inclusion and exclusion criteria. The minimum ulcer size of 3 mm, which might not satisfy some clinicians, is similar to that in almost all the studies of misoprostol. Despite their similarities, the 2 studies had some interesting demographic variations. Patients recruited by Yeomans and colleagues were more likely to be women, to have rheumatoid arthritis, to be infected with Helicobacter pylori, and to have a history of peptic ulceration.

Misoprostol used in the doses tested, 400 or 800 µg/d, has acid inhibitory effects. However, these effects do not explain its therapeutic effects, which are probably caused by the prostaglandin-mediated mucosal protective activity (known as cytoprotection). A key finding in both studies was that omeprazole, 20 mg/d, is as effective as 40 mg/d in healing NSAID ulcers. This suggests a ceiling beyond which no more acid is available to inhibit; therefore, doses of omeprazole > 20 mg/d offer no extra healing in long-term users of NSAIDs. This has cost-saving implications. It also explains the findings of Yeomans and colleagues of the superiority of omeprazole over the weaker acid inhibitor ranitidine in both healing ulcers and maintaining remission.

One should not assume, however, that acid explains the whole story in the multifactorial pathophysiology of NSAID-related gastroduodenal disease, particularly given the different patterns of activity of the two drugs, against erosions and ulcers. Omeprazole may be superior to misoprostol in maintaining remission only if the dose of misoprostol is divided to the maintenance level licensed at 400 µg/d. Omeprazole was better tolerated than misoprostol, but both studies showed a high rate of adverse events. However, adverse events are not necessarily the same as drug side effects, and it might not be possible to prove a causal relation, particularly in patients with chronic diseases. These important findings widen our choice of therapies used to heal NSAID-related ulcers and maintain their remission. They establish the role of omeprazole, 20 mg/d, and leave the door open for the use of other agents, particularly misoprostol.

Ali S. Taha, MD, PhD
Crosshouse HospitalKilmarnock, Scotland, UK