Upper gastroduodenal ulceration in arthritis patients treated with celecoxib.

OBJECTIVE To evaluate the comparative incidence of endoscopic gastroduodenal ulcers in patients with rheumatoid arthritis or osteoarthritis treated with celecoxib. DESIGN Quantitative systematic review of randomized controlled trials. SUBJECTS Patients (n = 4632) with rheumatoid arthritis or osteoarthritis reported in five trials. MAIN OUTCOME MEASURES Rate ratios, rate differences, and the number needed to harm were calculated for the incidence of endoscopically documented gastroduodenal ulcers. RESULTS Pooled rate ratios (RRs) relative to placebo for endoscopic ulcers at 12 weeks were 1.96 (95% CI 0.85 to 4.55) for celecoxib 100 mg twice daily and 2.35 (95% CI 1.02 to 5.38) for celecoxib 200 mg twice daily. There was no significant difference in gastroduodenal ulcers at 12 weeks between celecoxib 200 mg twice daily and celecoxib 100 mg twice daily; the corresponding pooled RR was 1.21 (95% CI 0.62 to 2.38). In contrast, celecoxib 200 mg twice daily was associated with a significantly lower rate of gastroduodenal ulcers than was naproxen 500 mg twice daily at 12 weeks (RR 0.24; 95% CI 0.17 to 0.33). On average, for every seven patients treated with naproxen, one more had an endoscopic ulcer than if they were treated with celecoxib. Celecoxib 200 mg twice daily also had a significantly lower risk of endoscopic ulcers than did either modified-release diclofenac 75 mg twice daily at 24 weeks (RR 0.24; 95% CI 0.11 to 0.52) or ibuprofen 800 mg three times daily at 12 weeks (RR 0.30; 95% CI 0.20 to 0.46). CONCLUSIONS Endoscopic studies have shown that celecoxib, at a wide range of doses, is associated with a lower incidence of gastroduodenal ulcers than are diclofenac, ibuprofen, or naproxen. The incidence rates of gastroduodenal ulcers associated with celecoxib were similar, although not equivalent, to placebo. Head-to-head comparisons suggest that, at the wide range of doses studied (100-800 mg/d), there are no dose-related increases in endoscopic gastroduodenal ulcers with celecoxib. The results of longer term comparative trials of celecoxib based on clinical outcomes are needed to determine celecoxib's ultimate risk-benefit profile.