Am Fam Physician. 2004;69(3):595-597
Searchable Question
How effective are selective cyclooxygenase-2 (COX-2) inhibitors in reducing the symptoms of rheumatoid arthritis when compared with nonselective nonsteroidal anti-inflammatory drugs (NSAIDs)?
Evidence-Based Answer
The efficacy of COX-2 inhibitors is similar to that of nonselective NSAIDs in reducing the symptoms of rheumatoid arthritis. [Strength of Recommendation: A]
Evidence Summary
Four randomized, double-blind, controlled trials1–4 compared the COX-2 inhibitors celecoxib (Celebrex), etoricoxib (Arcoxia), and rofecoxib (Vioxx) with the NSAIDs naproxen (Naprosyn) and diclofenac (Voltaren). Both classes were found to have similar efficacy in nearly all outcomes in patients with rheumatoid arthritis (see the table on the following page). [References 1 through 4—Evidence level 1B]
Only one study1 found significant differences in any outcomes. Patients randomized to the higher celecoxib dosage were more likely to be rated as improved on the patient's and physician's global assessments of disease activity than patients given naproxen (number needed to treat: approximately 10).
A study5 comparing celecoxib with naproxen in the treatment of rheumatoid arthritis or osteoarthritis found similar efficacy between the treatment groups; however, the results were not analyzed separately by arthritis type. [Evidence level 1B]
Recommendations from Others
The London-based National Institute for Clinical Excellence (NICE) recommends the use of COX-2 inhibitors over other NSAIDs only in patients with rheumatoid arthritis who have a high risk of serious gastrointestinal adverse effects.6 NICE recommends against routinely prescribing COX-2 inhibitors in patients with cardiovascular disease (because of a potentially increased risk of myocardial infarction) and patients taking low-dose aspirin (because the gastrointestinal protection of COX-2 inhibitors is reduced).
Clinical Commentary
Physicians can assure their patients with rheumatoid arthritis who are at increased risk of gastrointestinal ulcers that COX-2 inhibitors are as beneficial as nonselective NSAIDs in ameliorating their symptoms. However, there is no clinical reason to routinely prescribe COX-2 inhibitors to patients who do not have an increased risk of ulcers.
| Outcome | Trial, number of study participants, dosage comparison | ||||
|---|---|---|---|---|---|
| Collantes, et al., 20024 (n = 445) | Bombardier, et al., 20003 (n = 8,076) | Emery, et al., 19992 (n = 665) | Simon, et al., 19991 (n = 460) | Simon, et al., 19991 (n = 465) | |
| Etoricoxib (Arcoxia),* 90 mg once daily, versus naproxen (Naprosyn), 500 mg two times daily | Rofecoxib (Vioxx), 50 mg once daily, versus naproxen, 500 mg two times daily | Celecoxib (Celebrex), 200 mg two times daily, versus diclofenac SR (Voltaren SR), 75 mg two times daily | Celecoxib, 200 mg two times daily, versus naproxen, 500 mg two times daily | Celecoxib, 100 mg two times daily, versus naproxen, 500 mg two times daily | |
| Number of tender joints | – | Not assessed | – | – | – |
| Number of swollen joints | – | Not assessed | – | – | – |
| Patient's global assessment of pain | – | Not assessed | – | – | – |
| Functional disability | – | – | – | – | – |
| Patient's global assessment of disease activity | – | – | – | Better with celecoxib | – |
| Physician's global assessment of disease activity | – | – | – | Better with celecoxib | – |
| Percentage of patients meeting ACR-20 criteria† | – | Not assessed | – | – | – |
| Withdrawals for lack of efficacy | – | – | – | – | – |