Gastrointestinal tolerability of the selective cyclooxygenase-2 (COX-2) inhibitor rofecoxib compared with nonselective COX-1 and COX-2 inhibitors in osteoarthritis

Arch Intern Med. 2000 Oct 23;160(19):2998-3003. doi: 10.1001/archinte.160.19.2998.


Background: Most nonsteroidal anti-inflammatory drugs (NSAIDs) are nonselective cyclooxygenase (COX-1 and COX-2) inhibitors and are associated with a variety of upper gastrointestinal (GI) tract symptoms. The roles of COX-1 and COX-2 in the pathogenesis of these symptoms are unclear. To test whether COX-2 inhibition with rofecoxib would have greater GI tolerability than nonselective COX-1 and COX-2 inhibition, we compared the incidences of (1) treatment discontinuations for GI adverse events (AEs) and (2) prespecified dyspeptic-type GI AEs among patients with osteoarthritis treated with rofecoxib vs NSAIDs.

Methods: A prespecified, combined analysis of investigator-reported GI AEs in all 8 double-blind, randomized, phase 2b/3 osteoarthritis trials of rofecoxib was conducted. Patients included men and women with osteoarthritis (N = 5435); there was no upper age limit for entry. Treatments tested included rofecoxib, 12.5, 25, or 50 mg (combined), vs ibuprofen, diclofenac, or nabumetone (combined). Primary outcomes were the time (by survival analysis) to (1) treatment discontinuation due to GI AEs and (2) first reported dyspeptic-type GI AE. Between-treatment comparisons were made by log-rank test.

Results: The number of treatment discontinuations caused by GI AEs during 12 months was significantly lower (P=.02) with rofecoxib vs NSAIDs (8.2 vs 12.0 per 100 patient-years; relative risk, 0.70; 95% confidence interval, 0.52-0.94). The incidence of prespecified dyspeptic-type GI AEs during the first 6 months was significantly lower (P=.02) with rofecoxib vs NSAIDs (69.3 vs 85.2 per 100 patient-years; relative risk, 0.85; 95% confidence interval, 0.74-0.97). However, the difference between treatments in dyspeptic-type GI AEs was attenuated after 6 months.

Conclusion: Rofecoxib was associated with a lower incidence of treatment discontinuations due to GI AEs over 12 months and a lower incidence of dyspeptic-type GI AEs over 6 months than treatment with nonselective COX inhibitors, or NSAIDs. Arch Intern Med. 2000;160:2998-3003

Publication types

  • Meta-Analysis

MeSH terms

  • Abdominal Pain / chemically induced
  • Adult
  • Aged
  • Aged, 80 and over
  • Anti-Inflammatory Agents, Non-Steroidal / therapeutic use
  • Clinical Trials, Phase II as Topic
  • Clinical Trials, Phase III as Topic
  • Cyclooxygenase 1
  • Cyclooxygenase 2
  • Cyclooxygenase 2 Inhibitors
  • Cyclooxygenase Inhibitors / adverse effects*
  • Dyspepsia / chemically induced*
  • Female
  • Humans
  • Isoenzymes / antagonists & inhibitors
  • Lactones / adverse effects*
  • Male
  • Membrane Proteins
  • Middle Aged
  • Osteoarthritis / drug therapy*
  • Prostaglandin-Endoperoxide Synthases / metabolism*
  • Randomized Controlled Trials as Topic
  • Sulfones


  • Anti-Inflammatory Agents, Non-Steroidal
  • Cyclooxygenase 2 Inhibitors
  • Cyclooxygenase Inhibitors
  • Isoenzymes
  • Lactones
  • Membrane Proteins
  • Sulfones
  • rofecoxib
  • Cyclooxygenase 1
  • Cyclooxygenase 2
  • PTGS1 protein, human
  • PTGS2 protein, human
  • Prostaglandin-Endoperoxide Synthases