Article Text


FRI0035 Risk factors for clinically important upper gi events: the vigour study
  1. C Bombardier1,
  2. L Laine2,
  3. AS Reicin3,
  4. DR Shapiro4,
  5. RS Sperling3
  1. 1Institute for Work & Health, Mt. Sinai Hospital, Toronto, Canada
  2. 2GI Division, USCSM, Los Angeles
  3. 3Clin. Res
  4. 4Biostat., Merck & Co., Rahway, USA


Background The VIGOUR (VIOXX GI Outcomes Research) study was a prospective, randomised, double-blind comparison of rofecoxib (a COX-2 inhibitor) vs. naproxen (a dual COX-1/COX-2 inhibitor) in > 8000 patients with rheumatoid arthritis (RA). The study showed that rofecoxib was associated with significantly fewer clinically important upper gastrointestinal (GI) events (perforation, obstruction, bleeding, symptomatic gastric and duodenal ulcers) than naproxen.

Objectives To assess whether the risk reduction in clinically important upper GI events seen with rofecoxib was similar in “high risk” and “low risk” sub-populations; and, to predict the risk of these events based on baseline characteristics.

Methods In the VIGOUR trial, 8076 RA patients expected to require NSAIDs for > 1 yr were randomly assigned rofecoxib 50 mg qd or naproxen 500 mg bid for minimum of 6-month follow-up. Patients were classified as “low risk” or “high risk” for clinically important upper GI events based on recognised risk factors – age, steroid use, previous history of clinical GI events, and H. pylori sero-status. Relative risk analyses were conducted in both low- and high-risk subgroups. A multifactorial model to predict clinically important upper GI events was developed. Multiple steps in the analysis included: 1. Covariates were examined in a univariate Cox model; 2. Factors that trended (p < 0.20) plus treatment were put into Cox model and backwards stepwise regression chose significant (p = 0.05) predictors; and, 3. Interactions between each covariate with treatment were tested one at a time in Cox models.

Results The % risk reduction (rofecoxib relative to naproxen) of clinically important upper GI events was 88% in the low-risk subgroup, 51% in the high- risk subgroup and 54% in the overall population.

Abstract FRI0035 Table 1

Conclusion (1) The relative risk reduction of GI events in the rofecoxib group was maintained in both the high and low risk subgroups. (2) Multivariate analyses found the risk factors significantly associated with GI events to include age, prior history or clinical GI events (complicated and uncomplicated), disease severity, duration of disease, prior history of GI symptoms, prior use of low dose H2 blockers, steroids, and NSAIDs.

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.