Acetylsalicylic acid (aspirin) was synthesized some 100 years ago. Aspirin, and eventually the nonsteroidal anti-inflammatory drugs (NSAIDs), have provided considerable anti-inflammatory and analgesic benefit. All NSAIDs are, however, associated with a substantial level of adverse events, accounting for 20% and 25% of all reported adverse events in the UK and US, respectively [1,2]. These include dyspeptic symptoms, which are experienced by 15–40% of NSAID users , peptic ulceration in 10–30% of users and ulcer complications [4–6], which lead to hospitalization and are associated with mortality. It has been estimated that there are over 100 000 hospital admissions per year and 16 500 deaths per year due to NSAID-induced gastrointestinal complications in the US . Established risk factors for gastrointestinal complications, such as bleeding or perforation are: advanced age (> 60–65 years); a history of gastrointestinal bleeding; a history of peptic ulcer disease; multiple- or high-dose NSAIDs; and concomitant steroid treatment and concomitant anticoagulant treatment [7–11]. Serious comorbidity, such as cardiovascular disease [8,9] and severe rheumatoid arthritis [9,10], and a shorter duration of NSAID therapy have also been proposed as risk factors .
Despite their inherent problems, traditional NSAIDs are consumed on a daily basis by more than 30 million people worldwide, and they account for a vast number of prescriptions, around 20 million annually in the UK and 70 million annually in the US . There is a considerable economic burden associated with NSAID-associated gastro-duodenal damage. In the US, the adjusted mean annual cost of all types of medical care for gastrointestinal diseases in regular NSAID users is approximately double that in non-NSAID users; it is even higher in high-dose NSAID users . The economic burden might increase further in the future if NSAIDs become more widely used for the prevention of colon cancer and Alzheimer's disease, and if the use of low-dose aspirin for cardiovascular protection continues to increase.
However, the benefits of NSAIDs must be emphasized. Without continuous NSAID therapy, the symptoms of diseases such as rheumatoid arthritis would worsen and result in impairment of quality of life.
This article reviews the treatment options for healing and preventing gastro-duodenal injury associated with the use of non-selective NSAID treatment, and it poses the question whether such treatment is needed in the presence of cyclo-oxygenase-2 (COX-2)-selective NSAIDs.
Management of gastric and duodenal ulceration associated with NSAIDs
The H2-receptor antagonists have been used for some time to heal and prevent NSAID-associated ulcers; however, their efficacy is limited, particularly in healing ulcers during continued NSAID treatment and in the prevention of gastric ulcers.
Healing with H2-receptor antagonists
In a study of the healing effect of ranitidine on NSAID-associated ulcers (Table 1), patients were randomized to ranitidine (150 mg twice a day) and either continued or stopped NSAID treatment. At 8 weeks, both gastric and duodenal ulcers had healed in significantly more of those who had stopped NSAID treatment than in those who had continued the treatment .
Prevention with H2-receptor antagonists
Standard doses of ranitidine (150 mg twice a day for 8 weeks) and famotidine (20 mg twice a day for 24 weeks) reduced the incidence of duodenal ulceration, but not gastric ulceration, during continued NSAID use [15–17] (Table 2). Doubling the dose of famotidine to 40 mg twice a day, however, also significantly reduced the incidence of gastric ulceration .
The prostaglandin analogue misoprostol has been studied mainly in primary prophylaxis against NSAID-associated ulceration and has a dose-dependent protective effect against both duodenal and gastric ulcers. Misoprostol has also been shown to significantly reduce the incidence of severe upper gastrointestinal complications in a large study in older patients with rheumatoid arthritis . Its use has, however, been limited by its dose-dependent gastrointestinal side effects, such as diarrhoea and abdominal cramps .
Proton pump inhibitors
It has been shown in experimental studies in rats that the extent of gastric lesions caused by NSAIDs is highly dependent on intragastric pH, and that a pH of > 4.0 is needed to provide total protection against such lesions . Proton pump inhibitors (PPIs) provide more reliable 24-h suppression of acid secretion than either of the above drug classes [21,22] and therefore would seem to be a better treatment option for management of NSAID-associated duodenal, as well as gastric, ulceration.
Healing with omeprazole
Omeprazole (20 mg once a day) has been compared with ranitidine (150 mg twice a day) and misoprostol (200 μg four times a day) for healing of gastric and duodenal ulcers during continued NSAID treatment in two large multi-centre studies (Table 1). Healing rates were significantly higher with omeprazole at 4 weeks and 8 weeks than with either ranitidine or misoprostol [23,24].
In a gastric ulcer healing study by Walan et al. , ulcer healing rates were significantly higher with omeprazole (20 mg or 40 mg once a day) than with ranitidine (150 mg twice a day) in patients on continued NSAID treatment.
Bianchi Porro et al.  compared omeprazole (20 mg every morning) with sucralfate (2 g twice a day) for healing of gastric and duodenal ulcers and found that gastric ulcer healing rates were significantly higher with omeprazole than with sucralfate at both 4 weeks and 8 weeks. There was no significant difference in the duodenal ulcer healing rates between the two drugs.
Prevention with omeprazole
For prevention of gastric and duodenal ulceration, omeprazole (20 mg once a day) kept significantly more patients free of ulceration and gastrointestinal symptoms during 3–6 months’ treatment with continued NSAIDs than ranitidine (150 mg twice a day) , misoprostol (200 μg twice a day)  or placebo [24,27,28] (Table 2). Omeprazole was better tolerated than misoprostol, which was associated with diarrhoea in a significantly higher proportion of patients than omeprazole.
Healing with lansoprazole
The capacity of lansoprazole (15 mg and 30 mg once a day) and ranitidine (150 mg twice a day) to heal gastric ulcers during continued NSAID use has been compared in a large multi-centre study (Table 1). The percentage of patients with healed ulcers on both doses of lansoprazole was significantly higher than with ranitidine at 8 weeks .
Prevention with lansoprazole
Lansoprazole (15 mg and 30 mg once a day) was compared with misoprostol (200 μg four times a day) in the prevention of relapse in Helicobacter pylori-negative NSAID users with a history of gastric ulceration (Table 3). The proportion of patients remaining free from ulcers at 4 weeks, 8 weeks and 12 weeks was significantly higher on misoprostol and on both doses of lansoprazole compared with the placebo. There were no significant differences between the active treatment groups, except for the significantly higher incidence of diarrhoea reported by 22% of those on misoprostol compared with 3% of those on lansoprazole 15 mg, 7% of those on lansoprazole 30 mg and 3% of those on placebo .
The role of H. pylori in NSAID-associated ulcer disease is unclear. Studies with omeprazole have reported potentially beneficial effects of H. pylori infection in NSAID users, as demonstrated by a significantly increased likelihood of ulcer healing and significantly higher proportions of H. pylori-positive patients being kept in remission during continued NSAID treatment [23,24]. According to a study led by Hawkey, these findings might be explained by H. pylori stimulating the synthesis of mucosal prostaglandins . In contrast to Hawkey, Chan et al. have found that eradication of H. pylori does not impair the healing of NSAID-associated ulcers, and that H. pylori eradication before NSAID therapy reduces the occurrence of NSAID-induced ulcers [32–34].
Prophylaxis with either PPIs or misoprostol has been studied and used mainly in patients at risk of developing upper gastrointestinal complications during treatment with non-selective NSAIDs. It is reasonable to assume that COX-2-selective NSAIDs are being prescribed to a similar population. Administration of COX-2-selective NSAIDs to patients needing chronic treatment is likely to result in the co-prescription of low-dose aspirin for cardiovascular protection to up to 20% of this population, i.e. the same proportion as found among patients using a combination of celecoxib and low-dose aspirin in the celecoxib long-term arthritis safety study (CLASS) . The findings of fewer cases of acute myocardial infarction in patients treated with naproxen compared with rofecoxib in the VIGOR study  have further drawn attention to the need for low-dose aspirin in COX-2-selective NSAID users. The annualized incidence rates for gastrointestinal complications for patients in the CLASS study taking aspirin in a dose of ≤ 325 mg/day were 2.01% in celecoxib users and 2.12% in ibuprofen or diclofenac users (P = 0.92), showing that the combination of a COX-2-selective NSAID and aspirin in a low dose carries a risk of gastrointestinal complications similar to that of a non-selective NSAID. Hawkey et al. analysed potential risk factors for ulcer development over 12 weeks on placebo, rofecoxib 25 mg/day, rofecoxib 50 mg/day and ibuprofen 2400 mg/day and found that a history of peptic ulcer bleeding significantly increased the ulcer incidence across all treatment groups . This supports the contention that potential risk factors and, in some cases, prophylactic treatment should also be considered when prescribing COX-2-selective NSAIDs.
NSAID-associated upper gastrointestinal symptoms
Patients with chronic arthritis have a poor quality of life in comparison with a normal population because of impaired mobility, limitations in daily activities, pain, lack of energy and disturbed sleep . NSAID treatment can, on the one hand, markedly improve the quality of life in patients with arthritis by controlling pain and inflammation [39,40]. However, on the other hand, it can reverse this positive impact when patients suffer from upper gastrointestinal symptoms such as abdominal pain, heartburn or nausea associated with such treatment .
The proportion of patients on COX-2-selective NSAID treatment who report gastrointestinal symptoms such as dyspepsia, abdominal pain, epigastric discomfort, nausea and heartburn is comparable to that of patients on non-selective NSAIDs. Langman et al. have published a meta-analysis of eight studies in patients with osteoarthritis receiving rofecoxib (n = 3357), non-selective NSAIDs (n = 1564) or placebo (n = 514). The cumulative incidence of dyspeptic-type adverse events up to 6 months was 23.5% on rofecoxib and 25.5% on non-selective NSAIDs (P = 0.02), after which the incidence rates converged . Incidence rates for placebo were not given. In the CLASS study, 31.4% (n = 3987) reported gastrointestinal adverse effects on celecoxib, compared with 36.8% (n = 3981) on ibuprofen or diclofenac (P ≤0.05) .
Yeomans et al. and Hawkey et al. have studied whether upper gastrointestinal symptoms associated with NSAID treatment can be relieved with an H2-receptor antagonist, a PPI or misoprostol [23,24]. They found that non-selective NSAID users had significantly less abdominal pain and heartburn on omeprazole than on misoprostol, and the proportion of NSAID users showing an improvement in overall gastrointestinal symptoms was significantly greater on omeprazole than on ranitidine.
It seems reasonable to conclude that co-treatment with an H2-receptor antagonist has little to offer as regards the healing of ulcers during continued NSAID use or protection against gastric ulcers in particular. Either a PPI or misoprostol should be considered in NSAID users when any of the following conditions apply: previous gastrointestinal bleeding; previous gastric or duodenal ulceration; age > 60 years; serious co-morbidity; multiple- or high-dose NSAIDs; co-treatment with anticoagulants; co-treatment with steroids; bothersome upper gastrointestinal symptoms.
Effectiveness, tolerability and the costs associated with treatment should be taken into consideration when deciding whether misoprostol or a PPI should be chosen for co-treatment. Misoprostol has been shown to reduce the incidence of severe upper gastrointestinal complications. PPIs are documented not only for the prevention but also for the healing of ulcerations. PPIs are well tolerated and more effective than misoprostol for treatment of NSAID-associated gastrointestinal symptoms, as shown with omeprazole. Whether H. pylori should be eradicated in NSAID users and in patients about to start NSAID treatment is still not clear. Apart from the possible beneficial effects of H. pylori in NSAID users, there are good arguments to eradicate the bacterium in any case; this would probably also be preferred by most patients with an established H. pylori infection.
The experience with COX-2-selective NSAIDs is still limited. It remains to be seen, in clinical practice as well as in clinical studies, whether gastrointestinal problems associated with COX-2-selective NSAIDs really are less frequent than those associated with non-selective NSAIDs given alone or in combination with gastro-duodenal protection. The current experience, based on published data from clinical trials, is that upper gastrointestinal symptoms are still a significant clinical problem and that COX-2-selective NSAID users taking low-dose aspirin for cardiovascular protection have a similar risk of gastrointestinal complications as those taking a non-selective NSAID.
1. Wolfe F, Kleinheksel SM, Spitz PW, Lubeck DP, Fries JF, Young DY. et al
. A multicenter study of hospitalization in rheumatoid arthritis. Frequency, medical–surgical admissions, and charges. Arthritis Rheum 1986; 29: 614–619.
2. Hazelman BL. Incidence of gastropathy in destructive arthropathies. Scand J Rheumatol 1989; 78 (suppl) : 1–4.
3. Hirschowitz BI. Nonsteroidal antiinflammatory drugs and the gastrointestinal tract. Gastroenterologist 1994; 2: 207–223.
4. Griffin MR. Epidemiology of nonsteroidal anti-inflammatory drug-associated gastrointestinal injury. Am J Med 1998; 104 (suppl 3A) : 23S–29S.
5. Hawkey CJ. Non-steroidal anti-inflammatory drugs and peptic ulcers. BMJ 1990; 300: 278–284.
6. Myerson RM. NSAID
-associated gastroduodenal damage. J Pharm Med 1992; 2: 277–284.
7. Singh G. Recent considerations in nonsteroidal anti-inflammatory drug gastropathy. Am J Med 1998; 105 (suppl 1B) : 31S–38S.
8. Graham DY. Nonsteroidal anti-inflammatory drugs, Helicobacter pylori,
and ulcers: where we stand. Am J Gastroenterol 1996; 91: 2080–2086.
9. Simon LS, Hatoum HT, Bittman RM, Archambault WT, Polisson RP. Risk factors for serious nonsteroidal-induced gastrointestinal complications: regression analysis of the MUCOSA trial. Fam Med 1996; 28: 204–210.
10. Schoenfeld P, Kimmey MB, Scheiman J, Bjorkman D, Laine L. Nonsteroidal anti-inflammatory drug-associated gastrointestinal complications – guidelines for prevention and treatment [Review]. Aliment Pharmacol Ther 1999; 13: 1273–1285.
11. García Rodríguez LA, Jick H. Risk of upper gastrointestinal bleeding and perforation associated with individual non-steroidal anti-inflammatory drugs. Lancet 1994; 343: 769–772.
focus. Bandolier 1998; 5: 1–10.
13. Smalley WE, Griffin MR, Fought RL, Ray WA. Excess costs from gastrointestinal disease associated with nonsteroidal anti-inflammatory drugs. J Gen Intern Med 1996; 11: 461–469.
14. Lancaster-Smith MJ, Jaderberg ME, Jackson DA. Ranitidine in the treatment of non-steroidal anti-inflammatory drug associated gastric
ulcers. Gut 1991; 32: 252–255.
15. Ehsanullah RSB, Page MC, Tildesley G, Wood JR. Prevention of gastroduodenal damage induced by non-steroidal anti-inflammatory drugs: controlled trial of ranitidine. BMJ 1988; 297: 1017–1021.
16. Robinson MG, Griffin JW, Bowers J, Kogan FJ, Kogut DG, Lanza FL. et al
. Effect of ranitidine on gastroduodenal mucosal damage induced by nonsteroidal antiinflammatory drugs. Dig Dis Sci 1989; 34: 424–428.
17. Taha AS, Hudson N, Hawkey CJ, Swannell AJ, Trye PN, Cottrell J. et al
. Famotidine for the prevention of gastric
ulcers caused by nonsteroidal antiinflammatory drugs. N Engl J Med 1996; 334: 1435–1439.
18. Silverstein FE, Graham DY, Senior JR, Davies HW, Struthers BJ, Bittman RM. et al
reduces serious gastrointestinal complications in patients with rheumatoid arthritis receiving nonsteroidal anti-inflammatory drugs. Ann Intern Med 1995; 123: 241–249.
19. Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal toxicity of nonsteroidal antiinflammatory drugs. N Engl J Med 1999; 340: 1888–1899.
20. Elliott SL, Ferris RJ, Giraud AS, Cook GA, Skeljo MV, Yeomans ND. Indomethacine damage to rat gastric
mucosa is markedly dependent on luminal pH. Clin Exp Pharmacol Physiol 1996; 23: 432–434.
21. Jones DB, Howden CW, Burget DW, Kerr GD, Hunt RH. Acid suppression in duodenal ulcer
: a meta-analysis to define optimal dosing with antisecretory drugs. Gut 1987; 28: 1120–1127.
22. Chiverton SG, Burget DW, Salena BJ, Hunt RH. Does misoprostol
given as a single large dose improve its antisecretory effect? Aliment Pharmacol Ther 1989; 3: 403–407.
23. Yeomans ND, Tulassay Z, Juhász L, Rácz I, Howard JM, van Rensburg CJ. et al
. A comparison of omeprazole with ranitidine for ulcers associated with nonsteroidal antiinflammatory drugs. N Engl J Med 1998; 338: 719–726.
24. Hawkey CJ, Karrasch JA, Szczepañski L, Walker DG, Barkun A, Swannell AJ. et al
. Omeprazole compared with misoprostol
for ulcers associated with nonsteroidal antiinflammatory drugs. N Engl J Med 1998; 338: 727–734.
25. Walan A, Bader J-P, Classen M, Lamers CBHW, Piper DW, Rutgersson K. et al
. Effect of omeprazole and ranitidine on ulcer
healing and relapse rates in patients with benign gastric ulcer
. N Engl J Med 1989; 320: 69–75.
26. Bianchi Porro G, Lazzaroni M, Manzionna G, Petrillo M. Omeprazole and sucralfate in the treatment of NSAID
and duodenal ulcer
. Aliment Pharmacol Ther 1998; 12: 355–360.
27. Cullen D, Bardhan KD, Eisner M, Kogut DG, Peacock RA, Thomson JM. et al
. Primary gastroduodenal prophylaxis with omeprazole for non-steroidal anti-inflammatory drug users. Aliment Pharmacol Ther 1998; 12: 135–140.
28. Ekström P, Carling L, Wetterhus S, Wingren PE, Anker-Hansen O, Lundegårdh G. et al
. Prevention of peptic ulcer
and dyspeptic symptoms with omeprazole in patients receiving continuous non-steroidal anti-inflammatory drug therapy. A Nordic multicentre study. Scand J Gastroenterol 1996; 31: 753–758.
29. Agrawal NM, Campbell DR, Safdi MA, Lukasik NL, Huang B, Haber MM. Superiority of lansoprazole vs ranitidine in healing nonsteroidal anti-inflammatory drug-associated gastric
ulcers: results of a double-blind randomized, multicenter study. Arch Intern Med 2000; 160: 1455–1461.
30. Rose P, Huang B, Lukasik N, Collis C. Evidence that lansoprazole is effective in preventing NSAID
induced ulcers. Gastroenterology 1999; 116: A295.A295.
31. Hudson N, Balsitis M, Filipowicz F, Hawkey CJ. Effect of Helicobacter pylori
colonisation on gastric
mucosal eicosanoid synthesis in patients taking non-steroidal anti-inflammatory drugs. Gut 1993; 34: 748–751.
32. Chan FKL, Sung JJY, Suen R, Lee YT, Wu JCY, Leung WK. et al
. Does eradication of Helicobacter pylori
impair healing of nonsteroidal anti-inflammatory drug associated bleeding peptic ulcers? A prospective randomized study. Aliment Pharmacol Ther 1998; 12: 1201–1205.
33. Chan FKL, Sung JY, Chung SCS, To KF, Yung MY, Leung VKS. et al
. Randomised trial of eradication of Helicobacter pylori
before non-steroidal anti-inflammatory drug therapy to prevent peptic ulcers. Lancet 1997; 350: 975–979.
34. Hawkey CJ, Tulassay Z, Szczepanski L, van Rensburg CJ, Filipowicz-Sosnowska A, Lanas A. et al
. Randomised controlled trial of Helicobacter pylori
eradication in patients on non-steroidal anti-inflammatory drugs: HELP NSAIDs study. Lancet 1998; 352: 1016–1021.
35. Silverstein FE, Faich G, Goldstein JL, Simon LS, Pincus T, Whelton A. et al
. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis. The CLASS study: a randomized controlled trial. JAMA 2000; 284: 1247–1255.
36. Bombardier C, Laine L, Reicin A, Shapiro D, Burgos-Vargas R, Davis B. et al
., for the VIGOR Study Group. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med 2000; 343: 1520–1528.
37. Hawkey C, Harper S, Quan H, Bolognese J, Mortensen E. Effect of rofecoxib on endoscopic ulcers in osteoarthritis: patients analysis of potential risk factors [Abstract]. Ann Rheum Dis
38. Hunt SM, McEwen J, McKenna SP. Perceived health: age and sex comparisons in a community. J Epidemiol Comm Health 1984; 38: 156–160.
39. La Montagna G, Tirri G, Cacace E, Perpignano G, Covelli M, Pipitone V. et al
. Quality of life
assessment during six months of NSAID
treatment. Clin Exp Rheumatol 1998; 16: 49–54.
40. Fagnani F, Bouvenot G, Valat JP, Bardin T, Berdah L, Lafuma A. et al
. Medico-economic analysis of diacerein with or without standard therapy in the treatment of osteoarthritis. Pharmacoeconomics 1998; 13: 135–146.
41. Wolfe F, Kong SX, Watson DJ. Gastrointestinal symptoms and health related quality of life
in patients with arthritis. J Rheumatol 2000; 27: 1373–1378.
42. Langman MJ, Jensen DM, Watson DJ, Harper SE, Zhao P-L, Quan H. et al
. Adverse upper gastrointestinal effects of rofecoxib compared with NSAIDs. JAMA 1999; 282: 1929–1933.