Understanding Non-steroidal Anti-inflammatory Drugs (NSAIDs) and Selective COX-2 Inhibitors

What are nonselective NSAIDs and Selective COX-2 Inhibitors?

Non-steroidal anti-inflammatory drugs, often abbreviated NSAIDs, are drugs that reduce pain, fever and inflammation.  They are one of the most prescribed classes of medication worldwide with over 30 billion tablets purchased annually.  Approximately 70% of the elderly population takes NSAIDs each week.  COX-2 selective inhibitors are a type of NSAID that are also taken for relief of inflammation and pain.  They were developed because common NSAIDs were causing gastrointestinal problems for patients.  COX-2 selective inhibitors spare the gastrointestinal tract by virtue of their selectivity for an enzyme responsible for pain and inflammation, the cyclooxygenase-II enzyme, abbreviated COX-2.  Common NSAIDs inhibit both cyclooxygenase-I and cyclooxygenase-II. 

How do nonselective NSAIDs and Selective COX-2 Inhibitors Function?

Common or nonselective NSAIDs (such as aspirin, ibuprofen and naproxen) inhibit COX-1 and COX-2 enzymes.  COX-1 is expressed almost everywhere in the body and plays a critical role in protecting the gastrointestinal tract (GIT) from its own acidity by stimulating the production of prostaglandins.  COX-2 enzyme predominates in inflamed tissues and is not located in the stomach or small intestine. 

The nonselective inhibition of COX-1 and COX-2 by common NSAIDs reduces inflammation, but also causes a reduction in protective stomach mucosa, thereby making the GIT more susceptible to ulceration and bleeding.   

Why Were Selective COX-2 Inhibitors Developed?

Prior to 2003, patients with osteoarthritis and rheumatoid arthritis were almost exclusively given nonselective NSAIDs.  These enzyme inhibitors provided analgesic and anti-inflammatory benefits, but also inhibited the gastroprotective effects of COX-1.  In recent years, however, their widespread and often excessive use has resulted in an annual 103,000 hospitalizations and 16,500 deaths in the United States.  The major adverse drug reaction associated with nonselective NSAIDs is gastrointestinal irritation.  In response to the gastrointestinal toxicity of nonselective NSAIDs, two agents were developed: proton pump inhibitors and selective COX-2 inhibitors.  Proton pump inhibitors are a class of drugs that reduce the body’s production of gastric acid, thereby reducing ulceration and other GI problems.  Selective COX-2 inhibitors interact with only COX-2 enzymes and therefore do not interfere with gastric mucosa protection.  

Numerous clinical tests confirmed hopes that selective COX-2 inhibitors would produce less irritation on the gastrointestinal tract.  Patients and doctors noticed the difference, and the use of selective COX-2 inhibitors skyrocketed.  However, as we have seen with the highly-publicized withdrawal of Vioxx (rofecoxib) from the market, these selective inhibitors can also have serious side effects. 

NSAIDs and the Gastrointestinal Tract

Two large-scale studies comparing the adverse gastrointestinal effects of COX-2 specific inhibitors to those of traditional nonselective NSAIDs confirmed the relative safety of selective COX-2 inhibitors.  The 2000 Vioxx Gastrointestinal Outcomes Research Trial (VIGOR) involved a randomized double-blind comparison of 8076 patients with rheumatoid arthritis.[i]  Participants received either 50mg of rofecoxib or 500mg of naproxen twice daily for nine months.  The investigators found that significantly fewer adverse gastrointestinal events occurred with the rofecoxib treatment group.  Furthermore, both medications showed similar efficacy against rheumatoid arthritis.  A second study published in the same year also confirmed the relative gastrointestinal safety of selective COX-2 inhibitors, this time testing celecoxib.  The Celecoxib Long-Term Arthritis Safety Study Trial (CLASS) involved 8059 patients with osteoarthritis or rheumatoid arthritis in a prospective, randomized double-blind comparison of three groups.  Patients were assigned to receive either 400mg of celecoxib twice daily, 800mg of ibuprofen three times daily or 75mg of diclofenac twice daily.[ii]  Celecoxib was found to be associated with fewer upper gastrointestinal side effects, including symptomatic ulcers and ulcer complications. 

      These results seemed undeniable and each of these major trials was confirmed by smaller-scale follow-up studies. [iii]^,[iv]  Approximately two years later, the results of an enormous clinical study compared the frequency of adverse GI events for patients taking rofecoxib, celecoxib, nonselective NSAIDs or no drug therapy at all.[v]  This study by Mamdani et al[vi] corroborated the findings of the VIGOR and CLASS trials, but also reported a higher risk of GI events with rofecoxib than with celecoxib. 

COX-2 Selective Inhibitors are Milder on the GIT, but are Associated with Cardiovascular Risk

The excitement generated by the promising results of the VIGOR and CLASS trials overshadowed the side-effects of the selective COX-2 inhibitors.  Whereas the selective COX-2 inhibitors resulted in many fewer gastrointestinal problems, it also increased cardiovascular risk.  The VIGOR study itself reported a higher incidence of myocardial infarction in patients taking rofecoxib compared to those taking naproxen (0.4% vs 0.1%).[vii]  Further studies showed the relative cardiovascular risk for thrombotic event with rofecoxib to be twice as high compared with naproxen.[viii]  In the CLASS trial, however, there was no difference in cardiovascular events between the NSAID group and the celecoxib group.  A potential reason that may have led to lower cardiovascular events in the CLASS trial was that its duration was not long enough.  Data from the VIGOR trial shows a dramatic rise in cardiovascular events as treatment duration extends.  Another highly plausible reason for the relatively low cardiovascular events in the CLASS trial is that patients were allowed to take aspirin.[ix]  Aspirin’s blood thinning properties confer a significant cardioprotective effect to patients.  The third and most tangible reason that there was a difference between cardiovascular events in the CLASS and VIGOR trials is that the nature of the drugs might interact differently with the body.  In fact, a recently published study by the American College of Rheumatology suggests that low dose celecoxib is not associated with cardiovascular disease.[x]  However, multiple trials involving high-dose celecoxib have resulted in significant levels of cardiovascular events.[xi]^,[xii] 

What about Aspirin and Acetaminophen?

As suggested, one of the reasons for the lack of difference in cardiovascular events in the CLASS trial between celecoxib and the traditional NSAIDs is that patients were allowed to take low-dose aspirin while taking the selective COX-2 inhibitors.  Although aspirin may confer a cardioprotective advantage in addition to its anti-inflammatory properties, it is a traditional NSAID, and as such, has the ability to cause gastrointestinal problems. 

In 2000, the American College of Rheumatology recommended that acetaminophen be used as a first-line treatment for mild to moderate arthritic pain.[xiii]  This recommendation is echoed by the American Pain Society, which suggests that acetaminophen be used as the first-line treatment for mild pain.[xiv]  The refinement of this recommendation to exclude moderate pain is probably based on clinical trials that revealed a significant patient preference for NSAIDs instead of acetaminophen.  One such study from 2004 demonstrated better efficacy of NSAIDs, but also observed more withdrawals, reconfirming the gastrointestinal risks associated with this drug class.[xv]  It is important to remember, however, that acetaminophen is not without side effects.  Excessive use of acetaminophen has potential adverse effects including liver toxicity.

So What Should be Done Now?

Traditional NSAIDs, including aspirin, are associated with gastrointestinal problems, while selective COX-2 inhibitors are associated with cardiovascular complications.  Each patient’s unique risk profile and medical history is of the utmost importance in choosing the best treatment strategy.  For patients suffering from moderate to severe arthritis, the American College of Rheumatology recommends NSAIDs with a gastroprotective agent such as a proton pump inhibitor or a selective COX-2 inhibitor.[xvi]  Data suggests that a combination of a traditional NSAID with a PPI may cause as little harm to the gastrointestinal tract as seen with the selective COX-2 inhibitors.[xvii]  Although this has not yet been verified by rigorous large-scale trials, it is a promising finding.  In the meantime, talking with your doctor about your medical history and the available treatment options is the surest way to identify the most efficacious and safest treatment for you. 

Authors:

Eric J. Leroux, Princeton University, Princeton, NJ.

Juliana Khowong, M.D., New York-Presbyterian Hospital, The University Hospital of Columbia and Cornell, New York, NY

References