BACKGROUND: COX-2 specific inhibitors have been associated with a lower incidence of peptic ulcer disease and its complications (PUD) than non-specific NSAIDs. Point estimates from the CLASS study (up to 6 months) showed a relative risk of 0.59 (95% CI: 0.38–0.94) for celecoxib 1. Population-based epidemiological studies indicate that COX-2 specific inhibitors are prescribed to patients with a higher prevalence of risk factors for PUD. 1. Silverstein FE, Faich G, Goldstein JL et al. Gastrointestinal ToxicityWith Celecoxib vs NSAIDs Drugs for Osteoarthritis and Rheumatoid Arthritis: The CLASS Study: A Randomized Controlled Trial. JAMA 2000; 284(10): 1247–1255.
OBJECTIVE: To examine the effect of potential selective prescribing in the evaluation of the occurrence of spontaneous reports of PUD with celecoxib and non-specific NSAIDs.
METHODS: The FDA/Freedom of Information (FOI) database contains public information on selected fields from SRS/AERS. The SRS/AERS combined database, updated to the fourth quarter of 2002, was explored with QSCAN 2.1. We created a case definition for peptic ulcers and ulcer complications based on MedDRA preferred terms. Reports meeting the criteria were considered as cases. The proportional reporting ratio (PRR) was estimated as the OR for peptic ulcers and complications for celecoxib compared to non-specific NSAIDs (diclofenac sodium, etodolac, ibuprofen, and naproxen sodium).Multivariate logistic regression modelling was used to control for the effect of age, gender, and concomitant use of acetylsalicylic acid, warfarin, or gastroprotectors including misoprostol, sucralfate, H2 receptor antagonists, and proton pump inhibitors. Reports with concomitancy of various COX-2 specific inhibitors and/or nonspecific NSAIDs were excluded.
RESULTS: The proportion of reports of PUD was 17.8% for meloxicam, 17.7% for piroxicam, 7.7% for diclofenac, 6.2% for naproxen, 5.9% for celecoxib, 5.9% for ibuprofen and 4.9% for etodolac. The proportion of patients older than 64 years of age was higher for celecoxib (37.7%) compared to ibuprofen (19.2%). In addition, celecoxib showed the highest proportion of reports with co-medications like warfarin (5.2%), gastro protectors (16.8%) or aspirin (6.6%). The crude OR of PUD for celecoxib compared to non-specific NSAIDs was 0.94 (95% CI: 0.87–1.00). The OR decreased to 0.72 (95% CI: 0.67–0.78) after controlling for age and sex, and use of concomitant drugs.
CONCLUSION: Control of confounding led to point estimates of the proportional reporting ratio closer to the relative risk obtained in the CLASS study.
