Treating the pain associated with osteoarthritis is a common challenge for family physicians. Nonsteroidal anti-inflammatory drugs (NSAIDs) have long been used to manage this disease. The adverse side effect profile of NSAIDs is well known and includes gastrointestinal and renal complications. In patients who take NSAIDs regularly, upper endoscopic examinations have shown a 15 to 30 percent prevalence of ulcers in the stomach or duodenum.¹ The cyclooxygenase (COX) inhibitors, a new class of anti-inflammatory medications, have recently been introduced for use in the United States. NSAIDs inhibit COX, thereby inhibiting prostaglandin production. Two COX enzymes, COX-1 and COX-2, are known to be involved in prostaglandin synthesis. COX-1 generates prostaglandins that are involved in the protection of gastrointestinal mucosa, while COX-2 generates prostaglandins that mediate inflammation and pain in sites throughout the body. Selective COX-2 inhibitors may therefore relieve the pain of inflammation without deleterious effects on gastrointestinal mucosa.

Since their release, the COX-2 drugs have become very popular. Celecoxib (Celebrex), the first COX-2 inhibitor to reach the market, had sales totaling $1.5 billion in 1999. Up to 17.5 million prescriptions for this medication were written last year, eclipsing sildenafil (Viagra) as the fastest-selling new drug in history.² In this issue of American Family Physician, Noble and associates³ present an update on the use of this new class of medication. Their article makes the point that traditional NSAIDs and COX-2 inhibitors have equivalent analgesic efficacy, although few comparative studies are available. Therefore, the COX-2 inhibitors' advantage should be due to a more favorable side effect profile—specifically, their protection against gastrointestinal ulceration. The authors state that in patients with osteoarthritis and a high risk of ulceration (e.g., elderly patients with a history of gastrointestinal ulcer or bleeding), COX-2 inhibitors may be considered first-line therapy if therapy with acetaminophen is ineffective.

While short-term studies have shown that COX-2 inhibitors cause fewer gastrointestinal ulcers than NSAIDs, it is unclear how this difference translates in terms of clinically significant gastrointestinal bleeding. The rate of common adverse side effects such as dyspepsia is similar between the COX-2 inhibitors and NSAIDs. It is important for family physicians to put the COX-2 drugs in proper perspective. Acetaminophen (in dosages up to 4 g per day) is as effective as NSAIDs for the management of mild to moderate pain associated with osteoarthritis and is recommended as first-line therapy by the American College of Rheumatology.⁴ For localized chronic pain associated with osteoarthritis, topical capsaicin is also an effective analgesic. In patients who have chronic pain and an accompanying sleep disorder, concomitant use of low-dose sedating tricyclic antidepressants (e.g., 10 to 25 mg of amitriptyline [Elavil]) taken at bedtime should be considered. Patients may wish to consider the use of chondroitin and glucosamine.⁵ NSAIDs should be used in the lowest effective dosage.

In summary, COX-2 drugs offer an advantage in patients with a history of gastrointestinal bleeding who fail to respond to nonpharmacologic and non-NSAID therapy. This advantage benefits a small subset of the population of patients with osteoarthritis. Widespread use of COX-2 drugs is not justified.