Home

Current Issue

Archives

Guidelines for Authors

Classified Ads

Links

Search PubMed

Subscriptions

Subscriber Registration

Guidelines for Website Users

JRheum Update Service

Contact Info

NSAID Toxicity: Where Are We and How Do We Go Forward?

Nonsteroidal antiinflammatory drugs (NSAID) are the best recognized cause of iatrogenic pathology. They have been estimated to cause as many as 16,500 deaths per annum in the United States of America¹. That estimate is based upon an analysis of patients in the Arthritis, Rheumatism, and Aging Medical Information System (ARAMIS) database, which may not be representative of all patients using NSAID, and is much higher than other estimates².

Attempts to metaanalyze data on NSAID and ulcer complications, to produce as objective a quantitation as possible of the relationship between NSAID use and severe gastrointestinal (GI) complications, are therefore always welcome. The most recent of these, by Ofman and colleagues, is published in this issue of The Journal³. In their review of 2177 papers, they identified 55 published and 37 unpublished NSAID versus placebo randomized clinical trials (RCT), 57 case control studies, and 24 cohort studies. These yielded overall estimates of the odds/risk ratio for NSAID use that varied between 2.7 (cohort studies) and 5.36 (RCT). Interestingly, later epidemiological studies showed higher estimates of risk than earlier ones. This is probably because the definition of NSAID exposure became more tightly controlled as time progressed, whereas for RCT, exposure is controlled by trial protocol.

This suggests that epidemiological estimates of NSAID associated risk based on older studies may be conservative. Moreover, there may be other reasons why the direct effect of NSAID may be greater than previously estimated. This would occur if unmeasured factors causing ulcers were present in cases and controls but their confounding effects are not allowed for. Thus, the metaanalysis did not attempt to identify the interaction with other risk factors such as past history or Helicobacter pylori. Because risks associated with H. pylori would be present in both cases and controls, one would expect that an underestimate of the "pure" effects of NSAID would result. Recent publications suggest that this is the case. Stack, et al report an odds ratio associated with NSAID use of 11.3 (95% CI, 3.8-33.6) for ulcer bleeding in H. pylori negative patients⁴. In a metaanalysis of studies investigating the interaction between H. pylori and NSAID, Huang and colleagues reported an odds ratio of 18.1 (2.64-124) for endoscopic ulcers for NSAID use in the absence of H. pylori infection⁵ (although their estimates of risk for ulcer complications were very much lower). The 2 publications differ in their assessments of whether combinations of NSAID and H. pylori lead to further enhancement of risk [reduction to 7.8 (2.3-26.3) for ulcer complications in the study by Stack, et al and an increase to 61.1 (9.98-373) for endoscopy ulcers in the study of Huang and colleagues). Such wide discrepancies between studies suggest that the interaction between H. pylori and NSAID varies according to the population studied⁶.

Other combined analyses suggest that risk is dose dependent and may vary between individual drugs⁷. Non-dose dependent differences may occur to some extent because of differences in selectivity between the constitutive cyclooxygenase (COX)-1 and inducible COX-2 enzymes. The increased GI safety of COX-2 inhibitors compared to non-selective NSAID has been sufficiently striking for COX-2 inhibitors to be evaluated recently as a group by the United Kingdom National Institute for Clinical Excellence (NICE)⁸. NICE concluded that COX-2 inhibitors as a group were associated with fewer adverse events than non-selective NSAID, sufficient for their use to be recommended at least in high risk groups. Unfortunately NICE's evaluation did not define what a COX-2 inhibitor was. Drugs such as etodolac and meloxicam, which have a selectivity ratio of between 5 and 10 in whole blood assays, were included, but diclofenac, which consistently is COX-2 selective although to a somewhat lesser extent⁹, was not.

In their guidance, NICE did not draw distinctions between the safety of the individual drugs they analyzed. Nevertheless there are differences in the amount and robustness of data available (Table 1). The COX-2 hypothesis states that COX-2 inhibitors spare gastric mucosal prostaglandin synthesis and consequently cause no gastroduodenal injury. Until recently, data on gastric prostaglandin synthesis were available only for rofecoxib¹⁰ and etodolac^11,12, although the data on etodolac differ because they were not obtained using supratherapeutic doses (Table 1). Recently, an abstract has suggested that supratherapeutic doses of celecoxib also spare gastric mucosal prostaglandin synthesis¹³. A recent study of meloxicam 7.5 mg daily has shown reductions in prostaglandin synthesis that are similar to those seen with piroxicam 20 mg daily¹⁴.

Table 1. Comparison of effect on the GI system of 4 studied COX inhibitor.

As with biochemical mechanisms, so too the hypothesis of reduced injury has been tested robustly only with supratherapeutic doses and not been made in all cases (Table 1) with rofecoxib¹⁵ and celecoxib¹⁶. Both drugs have shown placebo levels of acute injury and chronic ulceration. Both have been subject to outcomes studies. These were positive in the case of rofecoxib but not significant in the case of celecoxib (almost certainly because of deficiencies in trial design rather than a failure of the COX-2 hypothesis). Shorter, less direct, uncontrolled or descriptive data with meloxicam and etodolac are consistent with reduced gastroduodenal damage, but the evidence is sketchier and a systematic evaluation of supratherapeutic doses has not been made.

One consequence of the attention COX-2 inhibitors have generated has been to draw attention to the non-GI toxicities of NSAID. In particular, both COX-2 inhibitors and NSAID result in fluid retention, edema, and hypertension. Where truly equivalent doses have been compared, the effects of different COX-2 inhibitors and NSAID have been similar. Where lower effective doses of one drug have been compared with another, not surprisingly there have been smaller changes in blood pressure. These considerations, along with evidence that GI toxicity is dose dependent⁷, emphasize the importance of using the lowest effective dose, whether a selective or non-selective NSAID is used. Perversely, much more attention has focused on coronary disease as a result of studies of high doses, with resulting inconsistency with other studies or with inappropriate choice of controls. Overall, studies of both selective and non-selective COX-2 inhibitors do not strongly suggest that these drugs have a direct effect on coronary thrombosis. Nevertheless, in due course evaluations of both selective and non-selective COX inhibitors, similar to those presented in this issue by Ofman and colleagues³, should be done. The challenge of such studies will be to assess the effect of drugs on overall health, morbidity, and mortality.

ACKNOWLEDGMENT

I thank Miss L.J. Garratt for her hard work in producing this manuscript.

Prof. CHRIS J. HAWKEY, DM, FRCP,
Division of Gastroenterology,
University Hospital Nottingham,
Queen's Medical Centre,
Nottingham, UK, NG7 2UH

Address reprint requests to Prof. Hawkey.

REFERENCES

Search PubMed for:

1. Singh G. Recent considerations in nonsteroidal anti-inflammatory drug gastropathy. Am J Med 1998;105:31S-38S. [MEDLINE]

2. Langman MJ. Risks of anti-inflammatory drug-associated damage. Inflamm Res 1999;48:236-8. [MEDLINE]

3. Ofman JJ, MacLean CH, Straus WL, et al. A metaanalysis of severe upper gastrointestinal complications of nonsteroidal antiinflammatory drugs. J Rheumatol 2002;29:804-12.

4. Stack WA, Atherton JC, Hawkey GM, Logan RFA, Hawkey CJ. Interactions between Helicobacter pylori and other risk factors for peptic ulcer bleeding. Aliment Pharmacol Ther 2002;In press.

5. Huang J-Q, Sridhar S, Hunt RH. Role of Helicobacter pylori infection and non-steroidal anti-inflammatory drugs in peptic ulcer disease: a meta analysis. Lancet 2002;359:14-22. [MEDLINE]

6. Hawkey CJ. Personal review: Helicobacter pylori, NSAIDs and cognitive dissonance. Aliment Pharmacol Ther 1999;13:695-702. [MEDLINE]

7. Henry D, Lim LL, Garcia Rodriguez LA, et al. Variability in risk of gastrointestinal complications with individual non-steroidal anti-inflammatory drugs: results of a collaborative meta-analysis. BMJ 1996;312:1563-6. [MEDLINE]

8. National Institute of Clinical Excellence. Guidance on the use of cyclo-oxygenase (COX) II selective inhibitors, celecoxib, rofecoxib, meloxicam and etodolac for osteoarthritis and rheumatoid arthritis. Technology Appraisal Guidance — No. 27. London: Government publication: 2001.

9. Warner TD, Giuliano F, Vojnovic I, Bukasa A, Mitchell JA, Vane JR. Nonsteroid drug selectivities for cyclo-oxygenase-1 rather than cyclo-oxygenase-2 are associated with human gastrointestinal toxicity: a full in vitro analysis [published erratum appears in Proc Natl Acad Sci USA 1999;96:9666]. Proc Natl Acad Sci USA 1999;96:7563-8. [MEDLINE]

10. Wight N, Gottesdiener K, Garlick NM, et al. Rofecoxib, a COX-2 inhibitor, does not inhibit human gastric mucosal prostaglandin production. Gastroenterology 2001;120:867-73. [MEDLINE]

11. Taha AS, McLaughlin S, Holland PJ, Kelly RW, Sturrock RD, Russell RI. Effect on gastric and duodenal mucosal prostaglandins of repeated intake of therapeutic doses of naproxen and etodolac in rheumatoid arthritis. Ann Rheum Dis 1990;49:354-8. [MEDLINE]

12. Laine L, Sloane R, Ferretti M, Cominelli F. A randomized, double-blind comparison of placebo, etodolac, and naproxen on gastrointestinal injury and prostaglandin production. Gastrointest Endosc 1995;42:428-433. [MEDLINE]

13. Lai KC, Cho CH, Hui WM, Hu W, Wong BCY, Lam SK. Lack of effect of celecoxib on gastromucosal prostaglandin synthesis. 9th European Gastroenterology Week. Abstract 2094.

14. Chang DM, Young TH, Hsu CT, et al. Endoscopic comparison of the gastroduodenal safety and the effects on arachidonic acid products between meloxicam and piroxicam in the treatment of osteoarthritis. Clin Rheumatol 2001:104-14. [MEDLINE]<

15. Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. VIGOR Study Group. N Engl J Med 2000;343:1520-8. [MEDLINE]

16. Silverstein FE, Faich G, Goldstein JL, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: A randomized controlled trial. Celecoxib Long-term Arthritis Safety Study. JAMA 2000;284:1247-55. [MEDLINE]

17. Simon LS, Lanza FL, Lipsky PE, et al. Preliminary study of the safety and efficacy of SC-58635, a novel cyclooxygenase 2 inhibitor. Efficacy and safety in two placebo-controlled trials in osteoarthritis and rheumatoid arthritis, and studies of gastrointestinal and platelet effects. Arthritis Rheum 1998;41:1591-602. [MEDLINE]

18. Lipscomb GR, Wallis N, Armstrong G, Rees WDW. Gastrointestinal tolerability of meloxicam and piroxicam: A double-blind placebo-controlled study. Br J Clin Pharmacol 1998;46:133-7. [MEDLINE]

19. Robinson MG, Griffin Jr JW, Bowers J, et al. Effect of ranitidine gastroduodenal mucosal damage induced by nonsteroidal antiinflammatory drugs. Dig Dis Sci 1989;34:424-8. [MEDLINE]

20. Lanza FL, Rack MF, Simon TJ, et al. Specific inhibition of cyclooxygenase-2 with MK-0966 is associated with less gastroduodenal damage than either aspirin or ibuprofen. Aliment Pharmacol Ther 1999;13:761-7. [MEDLINE]

21. Simon LS, Weaver AL, Graham DY, et al. Anti-inflammatory and upper gastrointestinal effects of celecoxib in rheumatoid arthritis: A randomized controlled trial. JAMA 1999;282:1921-8. [MEDLINE]

22. Patoia L, Santucci L, Furno P, et al. A 4-week, double-blind, parallel-group study to compare the gastrointestinal effects of meloxicam 7.5 mg, meloxicam 15 mg, piroxicam 20 mg and placebo by means of faecal blood loss, endoscopy and symptom evaluation in healthy volunteers. Br J Rheumatol 1996;35:61-7. [MEDLINE]

23. Lipscomb GR, Wallis N, Armstrong G, Rees WD. Gastrointestinal tolerability of meloxicam and piroxicam: a double-blind placebo-controlled study. Br J Clin Pharmacol 1998;46:133-7. [MEDLINE]

24. Lanza F, Rack MF, Lynn M, et al. An endoscopic comparison of the effects of etodolac, indomethacin, ibuprofen, naproxen, and placebo on the gastrointestinal mucosa. J Rheumatol 1987;14:338-41. [MEDLINE]

25. Russell RI. Endoscopic evaluation of etodolac and naproxen, and their relative effects on gastric and duodenal prostaglandins. Rheumatol Int 1990;10:17-21. [MEDLINE]

26. Lipscomb GR, Wallis N, Armstrong G, Goodman MJ, Rees WDW. Gastric mucosal adaptation to etodolac and naproxen. Aliment Pharmacol Ther 1995;9:379-85. [MEDLINE]

27. Bianchi Porro G, Caruso I, Petrillo M, Montrone F, Ardizzone S. A double-blind gastroscopic evaluation of the effects of etodolac and naproxen on the gastrointestinal mucosa of rheumatic patients. J Intern Med 1991;229:5-8. [MEDLINE]

28. Hawkey C, Laine L, Simon T, et al. Comparison of the effect of rofecoxib (a cyclooxygenase 2 inhibitor), ibuprofen, and placebo on the gastroduodenal mucosa of patients with osteoarthritis: A randomized, double-blind, placebo-controlled trial. Arthritis Rheum 2000;43:370-7. [MEDLINE]

29. Hawkey C, Kahan A, Steinbruck K, et al. Gastrointestinal tolerability of meloxicam compared to diclofenac in osteoarthritis patients. International MELISSA Study Group. Meloxicam Large-scale International Study Safety Assessment. Br J Rheumatol 1998;37:937-45. [MEDLINE]

30. Dequeker J, Hawkey C, Kahan A, et al. Improvement in gastrointestinal tolerability of the selective cyclooxygenase (COX)-2 inhibitor, meloxicam, compared with piroxicam: Results of the safety and efficacy large-scale evaluation of COX-inhibiting therapies (SELECT) trial in osteoarthritis. Br J Rheumatol 1998;37:946-51. [MEDLINE]

31. Neustadt DH. Double blind evaluation of the longterm effects of etodolac versus ibuprofen in patients with rheumatoid arthritis. J Rheumatol 1997;24:17-22. [MEDLINE]