The recently released cyclooxygenase-2 (COX-2) inhibitors are stated to have a lower incidence of gastrointestinal (GI) side effects than nonsteroidal anti-inflammatory drugs (NSAIDs), which interfere with COX-1 and -2. Specifically, NSAIDs have been shown in other studies to be associated with a three- to 10-fold higher risk of GI injury and/or death. Simon and colleagues conducted this randomized controlled trial to compare celecoxib with naproxen or placebo in patients with rheumatoid arthritis. The goal was to determine if celecoxib was as efficacious as traditional NSAIDs in terms of anti-inflammatory and analgesic effects without causing as much GI damage.

Patients who met the American College of Rheumatology criteria for rheumatoid arthritis were included if their arthritis therapy was stable. Patients were excluded if they had active GI tract disease or a history of malignancy within the past five years or a peptic ulcer within the past 30 days. If the baseline endoscopy revealed esophageal or peptic ulceration or more than 10 erosions, they were also excluded. Patients with a history of peptic ulcer disease were not excluded. A baseline test for Helicobacter pylori infection was performed, along with a history and physical examination. A global assessment of the arthritis activity was made by the patients and the physicians.

Analgesics and NSAIDs were discontinued during a two- to seven-day washout period. Within the week before beginning the study medication, an upper GI endoscopy was done to assess for petechiae, erosions and ulcerations. Patients were randomly assigned to receive placebo, celecoxib (in dosages of 100, 200 or 400 mg twice daily) or naproxen (in a dosage of 500 mg twice daily). Patients were allowed to continue taking oral glucocorticoids or disease-modifying antirheumatic drugs but were not allowed to take NSAIDs, anticoagulants or injectable corticosteroids. Efficacy and safety assessments were performed at weeks 2, 6 and 12. Adverse reactions were recorded. A second endoscopy and a second H. pylori test were performed at the end of the study.

A total of 1,149 patients enrolled in the study. The groups did not have significantly different endoscopy scores or physical assessments before beginning the study. Celecoxib (all dosages) and naproxen provided significant improvement in rheumatoid arthritis signs and symptoms. In the physicians' global assessment, celecoxib performed significantly better than placebo but, at 12 weeks, naproxen was not found to be better than placebo in terms of global assessment of functioning. Pain and morning stiffness were judged by the patients to be significantly better with celecoxib and naproxen when compared with placebo.

GI ulcers developed in 4 percent of patients receiving placebo compared with 6 percent taking the 100-mg dosage of celecoxib, 4 percent taking the 200-mg dosage, 6 percent taking the 400-mg dosage and 26 percent of patients taking naproxen. Adverse events experienced by those in the celecoxib group included headache, dyspepsia and abdominal pain. GI adverse events occurred in 19 percent of patients taking placebo, 28, 25 and 26 percent for those taking the 100-, 200- and 400-mg dosages of celecoxib, respectively, and in 31 percent of those taking naproxen.

The authors conclude that celecoxib is safe and efficacious in the treatment of rheumatoid arthritis. In this study, the incidence of gastroduodenal ulcers is less in those taking celecoxib than in those taking a traditional NSAID (naproxen). In a related editorial, Peterson and Cryer caution that although endoscopic ulcers occurred at a much higher incidence in the naproxen group than in the celecoxib group, they were so small as to be clinically unimportant. Until larger studies show that COX-2 inhibitors are cost-effective, they suggest moderate enthusiasm for the use of these agents.