Skip to main content Back to Top


Newer, Costlier NSAIDs Still Not for Everyone

Cheryl A. Thompson

As clinicians debate the benefits of using nonsteroidal anti-inflammatory drugs (NSAIDs) that inhibit a specific form of cyclooxygenase (COX), new research fuels the discussion about adverse effects but fails to show that these new agents spare the gastrointestinal tract from serious damage.

One of these studies, in the Nov. 24 issue of the Journal of the American Medical Association (JAMA), involved 688 patients with a diagnosis and symptoms of rheumatoid arthritis. The patients completed 12 weeks of treatment with celecoxib, which specifically inhibits cyclooxygenase-2 (COX-2), naprosyn, or a placebo. 

Physicians, using an endoscope to examine the lining of the stomach and duodenum, found ulcers in 24 percent of the patients in the naprosyn group. Only 5 percent of the patients in the celecoxib group had ulcers—about the same as with the placebo group. Because ulcers are generally believed by the medical community to be precursors for severe complications, the researchers concluded that COX-2-inhibiting NSAIDs might produce fewer of these serious events than might nonselective agents such as naprosyn. 

The second study, also in the Nov. 24 issue of JAMA, analyzed eight trials that compared the use of rofecoxib, nonselective NSAIDs, and placebo for six weeks to one year by 5,435 patients with osteoarthritis. From a committee's review of the studies, the researchers reported a significantly lower incidence of gastrointestinal ulcers, perforations, and bleeding in the patients who had received rofecoxib instead of a nonselective NSAID. 

In the editorial accompanying the two research articles, Walter L. Peterson, M.D., and Byron Cryer, M.D., at the Department of Veterans Affairs Medical Center in Dallas noted that the ulcers detected by endoscopic examination in the celecoxib study were clinically unimportant. Because NSAIDs so rarely cause major ulcers, Peterson and Cryer wrote, the researchers must study more patients in order to determine whether celecoxib causes this adverse effect less often than do traditional agents. 

From their analysis of costs and benefits, Peterson and Cryer decided that the use of COX-2-inhibiting NSAIDs might be justified for patients at high risk for ulcer complications, such as people 75 years or older who have had an ulcer or gastrointestinal bleeding. "Costs strongly favor using a generic, nonselective NSAID in patients at low risk for ulcer complications," they concluded.