Article Text


THU0224 Celecoxib decreased dyspepsia in osteoarthritis and rheumatoid arthritis patients: severity of dyspepsia assessment (soda) results from the ?celecoxib long-term arthritis safety study? (class 2), a randomised trial comparing celecoxib and diclofenac
  1. JL Goldstein1,
  2. BM Pena2,
  3. D Dedhiya3,
  4. LS Simon4
  1. 1Department of Medicine, University of Illinois at Chicago, Chicago, USA
  2. 2Research and Development, Pfizer, Inc., New York, USA
  3. 3Research and Development, Pharmacia Corporation, Skokie, USA
  4. 4Department of Medicine, Beth Israel Deaconess Medical Center, Boston, USA


Background Dyspepsia associated with NSAIDs is a cost-driver in the clinical management of arthritis patients and leads to discontinuation of medications.

Objectives To assess the impact of dyspepsia in arthritis patients randomised to celecoxib (C) or diclofenac (D) using SODA.

Methods As part of the double-blind, randomised CLASS trial, SODA (previously validated in dyspepsia patients) was administered at baseline and Weeks 4, 13, 26, and 52 (final visit) to 1997 C patients (400 mg BID, 4 x the OA dose) and 1996 D patients (75 mg BID). Patients with > = 1 response missing at baseline were excluded, leaving 1791 C and 1816 D patients in the modified ITT (M-ITT) group. SODA has three subscales: pain intensity (6 items characterising the abdominal pain associated with dyspepsia, ranging from 2 to 47); nonpain symptoms (7 items describing the impact of dyspeptic symptoms on daily living activities, ranging from 7 to 35); and satisfaction (4 items describing satisfaction with dyspepsia Rx). Lower SODA scores are better.

Results SODA change scores were moderately correlated with number of dyspepsia adverse events and increased with dyspepsia adverse event severity levels (none, mild, moderate, severe) and dyspepsia withdrawal status, irrespective of treatment. An M-ITT analysis with last observation carried forward showed that both groups were equivalent at baseline. Patients receiving C had lower pain intensity at all follow-up time points, ranging from 12.3 vs 14.5 at Week 4 to 12.6 vs 14.2 at Week 52 (p < 0.001 at all time points). Patients receiving C were more satisfied at all follow-up time points, ranging from 6.0 vs 6.7 at Week 4 to 6.3 vs 6.9 at Week 52 (p < 0.001 for all time points). Patients receiving C had fewer nonpain symptoms at Week 4 (12.0 vs 12.4, p < 0.003) and showed a trend toward fewer nonpain symptoms at subsequent time points. Self-reported dyspepsia was lower with C (16.5% vs 19.5%, p < 0.05). Similarly, GI-driven AE withdrawal rates were significantly lower for C than D at Week 52 (12.2% vs 16.6%, p < 0.05).

Conclusion Celecoxib patients had less pain intensity and were more satisfied than diclofenac patients, as measured by SODA. These results parallel self-reported dyspepsia and GI-driven withdrawal rates in the study.

Sponsored by Pharmacia Corporation and Pfizer, Inc.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.