This site is intended for health professionals only

Coxibs: overview of safety in osteoarthritis

MT Nurmohamed
Rheumatologist, Epidemiologist and Clinical investigator
Department of Rheumatology
VU Medical Center and the Jan van Breemen Institute Amsterdam
The Netherlands
E:[email protected]

The management of osteoarthritis (OA) includes nonpharmacological measures such as patient education, weight loss and physical therapy as well as drug therapy.(1) Acetaminophen is comparably as effective as nonsteroidal anti-inflammatory drugs (NSAIDs) in patients with mild-to- moderate pain, whereas for severe pain NSAIDs are more effective than acetaminophen.(2,3)

Traditional NSAIDs inhibit both cyclooxygenase-1 (COX-1) and cyclooxygenase-2 (COX-2). COX-1 is constitutively expressed in platelets and gastroduodenal mucosa, while COX-2 is upregulated in inflammatory processes. The clinical efficacy of conventional NSAIDs results from their COX-2 inhibitory effect, whereas their gastrointestinal side-effects are thought to be related to their COX-1 inhibition. Hence, COX-2 selective inhibitors (coxibs) were developed.

Presently, several selective coxibs are available: celecoxib, rofecoxib, valdecoxib and etoricoxib. Coxibs may have a role in patients with an increased risk of serious gastrointestinal adverse events, such as bleeding, perforation and obstruction. These risk factors include:

  • Age above 65 years.
  • Use of corticosteroids.
  • History of peptic ulcer disease and of upper gastrointestinal bleeding.
  • Use of anticoagulation.(1,4)

Gastrointestinal safety compared with conventional NSAIDS
There is mounting evidence that the use of celecoxib and rofecoxib results in a significant decrease in gastrointestinal adverse events compared with the conventional NSAIDs.

The largest trials published so far are the Vioxx Gastrointestinal Outcomes Research (VIGOR) trial and the Celecoxib Long-term Arthritis Safety Study (CLASS) investigation.

In the VIGOR trial,(5) rofecoxib 50mg once daily was compared with 500mg naproxen twice daily in more than 8,000 rheumatoid arthritis (RA) patients. The (past) use of aspirin was an exclusion criterion. Patients were treated for nine months, and the incidence of clinically relevant upper gastrointestinal side-effects was 2.1 per 100 patient-years in the rofecoxib-treated patients and 4.5 per 100 patient-years in the naproxen-treated patients (relative risk 0.5; 95% confidence interval [CI] 0.3–0.6; p<0.001).

The CLASS trial compared celecoxib 400mg twice daily with either ibuprofen 800mg three times daily or diclofenac 75mg twice daily.(6) More than 8,000 patients with either OA or RA with an intended treatment period of at least six months were investigated. Aspirin was allowed, and actually one-fifth of the patients used aspirin. There was no statistical difference in the primary endpoint of ulcer complications, with an incidence rate of 0.8% per year in the celecoxib-treated patients versus 1.5% in either NSAID group (p=0.09). When combined with symptomatic ulcers these rates were 2.1 and 3.5 respectively (p=0.02).

Very recently, a double-blind placebo-controlled parallel-group study in 1,052 OA patients compared the incidence of endoscopic-proven gastroduodenal ulcers in patients treated with 10 or 20mg valdecoxib once daily, ibuprofen 800mg three times daily or diclofenac twice daily.(7) Efficacy responses were comparable between the four active treatment groups. The incidence of gastroduodenal ulcers at 12 weeks was 5% in patients treated with 10mg valdecoxib, 4% in the 20mg valdecoxib group, 7% in the placebo group, 16% in the ibuprofen-treated patients, and 17% in the patients receiving diclofenac. This trial confirms the better gastrointestinal safety profile of coxibs compared with conventional NSAIDs.

Gastrointestinal safety of rofecoxib compared with celecoxib

Although the rate of upper gastrointestinal bleeding for the coxibs is significantly lower than that observed with conventional NSAIDS, there might be a small but statistically significant difference in gastrointestinal haemorrhage between celecoxib and rofecoxib, in favour of celecoxib, as recently suggested by a retrospective population-based study.(8) However, this study is hampered by its design, and the matter will be settled only by an appropriate comparative trial that also addresses the risk–benefit ratio of the two coxibs.

Cardiovascular safety compared with conventional NSAIDS

The rate of cardiovascular events measured in the VIGOR trial was 0.8% in the rofecoxib group and 0.4% in the naproxen group.(5) This difference was largely caused by a difference in myocardial infarction and occurred predominantly in patients at high risk of cardiovascular disease for whom low-dose aspirin was indicated but not given. The relative risk among these patients between the rofecoxib recipients and naproxen recipients was 4.89 (95% CI 1.4–16.9; p=0.01), and in the patients not needing aspirin the risk was 1.89 (95% CI 1.0–3.5; p=0.04).(9)

The possible increase in cardiovascular events induced by rofecoxib caused a continuing debate as to whether or not coxibs are associated with an enhanced cardiovascular risk. Theoretically, the production of thromboxane without the simultaneous formation of prostacyclin due to the COX-2 inhibition could cause a procoagulant state.

Article continues below this sponsored advert
Cogora InRead Image
Explore the latest advances in clinical care at events delivered by renowned experts from CofE

However, in the CLASS trial,(6) the rate of cardiovascular events was found to be similar in both treatment groups. This difference with the VIGOR trial might indicate that the induction of cardiovascular events by coxibs was masked by aspirin use in the CLASS trial. Alternatively, naproxen might be cardioprotective,(5) although sound evidence for this interesting hypothesis is lacking. None of the nonaspirin NSAIDs has been adequately investigated in terms of their potential (preventive) cardiovascular effects.(10)

Finally, the cardiovascular risk in RA patients might be increased compared with the general population.(11) The VIGOR patients were RA patients, whereas mainly OA patients were studied in the CLASS investigation. OA patients have a cardiovascular risk comparable with the normal population, and this lower “background” cardiovascular risk might be an alternative explanation for the observed difference between the two studies.

Overall, there is no firm evidence that coxibs result in an increased rate of cardiovascular events when they are used instead of conventional NSAIDs. However, there is a need for powered trials to address these topics, and such trials are presently being conducted.

Celecoxib-treated patients experienced less hypertension than patients on other NSAIDs in the CLASS trial – 1.7% versus 2.3% respectively (p<0.05) – whereas more rofecoxib patients than naproxen patients developed hypertension in the VIGOR trial. In the rofexocib group the mean increase in systolic blood pressure was 4.6mmHg with a diastolic increase of 1.7mmHg; for the naproxen group these data were 1.0mmHg and 0.1mmHg respectively.(9)

Cardiovascular safety of rofecoxib compared with celecoxib
Rofecoxib and celecoxib were recently compared in a six-week double-blind trial of 810 elderly OA patients using antihypertensive agents. Patients were treated with either 25mg rofecoxib (n=399) or 200mg celecoxib (n=411).(12) Systolic blood pressure increased significantly in 17% of rofecoxib- and 11% of celecoxib-treated patients respectively (p=0.03). Diastolic blood pressure increased in 2.3% in the rofecoxib group and 1.5% in the celecoxib group.

Oedema was observed in 9.5% of the rofecoxib patients and 4.9% of the celecoxib patients. The clinical significance of the observed oedema is unclear as another study indicated that the oedema disappeared in many of the rofecoxib patients when this drug was continued.(13) There is discussion about whether or not equipotent dosages were investigated with respect to efficacy, as there are some suggestions that the efficacy of the investigated dose of celecoxib was lower  compared with the studied rofecoxib dose.(14)

It is still unclear whether hypertension and oedema occur less frequently in coxibs compared with traditional NSAIDs and whether or not there are intrinsic differences between celecoxib and rofecoxib.

In general, coxibs appear to have comparable efficacy with conventional NSAIDs, with a better gastrointestinal safety profile. This might be useful for OA patients who carry an enhanced risk of gastrointestinal side-effects. Alternatively, the combination of a conventional NSAID with a proton pump inhibitor or misoprostol could be considered.

It is still not known whether or not coxibs are associated with an enhanced cardiovascular risk, and it was suggested very recently that this risk might in fact be dose-dependent. (15) It is hoped that the investigations that are currently underway will help to resolve these issues.


  1. American College of Rheumatology Subcommittee on Osteoarthritis Guidelines. Recommendations for the medical management of osteoarthritis of the hip and knee: 2000 update. Arthritis Rheum 2000;43:1905-15.
  2. Felson DT. The verdict favors nonsteroidal antiinflammatory drugs for treatment of osteoarthritis and a plea for more evidence on other treatments. Arthritis Rheum 2001;44:1477-80.
  3. Pincus T, Koch GG, Sokka T, et al. A randomised, double-blind, crossover clinical trial of diclofenac plus misoprostol versus acetaminophen in patients with osteoarthritis of the hip or knee. Arthritis Rheum 2001;44:1587-98.
  4. Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal toxicity of non-steroidal anti-inflammatory drugs. N Engl J Med 1999;340:1888-99.
  5. Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med 2000;343:1520-8.
  6. Silverstein FE, Percival MD, Brideau C, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: a randomised controlled trial. JAMA 2000;284:1247-5.
  7. Sikes DH, Agrawal NM, Zhao WW, Kent JD, Recker DP, Verburg KM. Incidence of gastroduodenal ulcers associated with valdecoxib compared with that of ibuprofen and diclofenac in patients with osteoarthritis. Eur J Gastroenterol Hepatol 2002;14:1101-11.
  8. Mamdani M, Rochon PA, Juurlink DN, et al. Observational study of upper gastrointestinal haemorrhage in elderly patients given selective cyclo-oxygenase-2 inhibitors or conventional non-steroidal anti-inflammatory drugs. BMJ 2002;325:624-9.
  9. Mukherjee D, Nissen SE, Topol EJ. Risk of cardiovascular events associated with selective COX-2 inhibitors. JAMA 2001;286:954-9.
  10. Patrono C, Coller B, Dalen JE, et al. Platelet active drugs: the relationships among dose, effectiveness, and side effects. Chest 2001;119:S39-63.
  11. Van Doornum S, McColl G, Wicks IP. Accelerated atherosclerosis: an extraarticular feature of rheumatoid arthritis. Arthritis Rheum 2002;46:862-73.
  12. Whelton A, Fort JG, Puma JA, et al. Cyclooxygenase-2-specific inhibitors and cardiorenal function: a randomized, controlled trial of celecoxib and rofecoxib in older hypertensive osteoarthritis patients. Am J Ther 2001;8:85-95.
  13. Schnitzer TJ. Cyclooxygenase-2-specific inhibitors: are they safe? Am J Med 2001;110(1A):S46-9.
  14. Geba GP, Weaver AL, Schnitzer TJ, et al. A clinical trial comparing rofecoxib to celecoxib and acetaminophen in the treatment of osteoarthritis (OA): early efficacy results [abstract]. Ann Rheum Dis 2000;59 Suppl 1:291.
  15. Ray WA, Stein CM, Daugherty JR, Hall K, Arbogast PG, Griffin MR. Cox-2 selective non-steroidal anti-inflammatory drugs and risk of serious coronary heart disease. Lancet 2002;360:1071-3.

Be in the know
Subscribe to Hospital Pharmacy Europe newsletter and magazine