BACKGROUND: Nonsteroidal anti-inflammatory agents are associated with a high incidence of gastrointestinal side effects due to nonspecific inhibition of cyclo-oxygenase (COX) enzymes, COX-1 and 2. Selective inhibition of COX-2 would minimize the potential for gastrointestinal toxicity without compromising efficacy. This study was conducted to investigate 1) the clinical efficacy and 2) the safety of meloxicam, a preferential inhibitor of COX-2, compared with piroxicam in chronic osteoarthritis patients. METHODS: This was designed as a randomized, open labeled, multi-clinic prospective trial. Patients received either meloxicam 7.5 mg or piroxicam 10 mg twice daily for two weeks. Parameters measured were demographic data, disease characteristics, static and dynamic visual analog scale scores (VAS score: 0 = no pain, 10 = extremely severe pain), side effects and their incidences, weekly drop out rates, global efficacy assessed by patients and physicians (1 = extremely satisfactory, 2 = satisfactory, 3 = unsatisfactory, 4 = extremely unsatisfactory), and the changes of disease status assessed by the patients (1 = remarkably improved, 2 = improved, 3 = no change, 4 = worse). RESULTS: There were no significant differences between the two groups in respect to demographic data and disease status. VAS scores at one week showed significant decrease from the baseline, but had no further improvement when compared with those at two weeks. These trends were similar in both groups. At the first week, the meloxicam group had lower incidences of side effects, but at the end of the study, these numbers were compatible between the two groups. Global efficacy evaluated by patients and physicians were highly satisfactory. The changes of disease status assessed by patients also revealed good improvements in both groups. Conclusion: Meloxicam had an analgesic effect for chronic osteoarthritis comparable to piroxicam. It appeared to have a better safety profile for short term therapy, but not in long term.