COX-2 Inhibitors and Cardiovascular Disease: Reason for Concern?
Abstract & Commentary
Synopsis: COX-2 inhibitors have recently come under scrutiny for an apparent increased risk of serious cardiovascular events associated with their use. A careful assessment of the present data supports the view that the cardiovascular risk profile is likely no greater than most standard NSAIDs, and no significant change in the use pattern of these drugs is presently warranted.
Source: Mukherjee D, et al. JAMA. 2001;286(8):954-959.
This paper is a retrospective analysis of available data on the potential cardiovascular risks associated with the use of selective cyclooxygenase-2 (COX-2) inhibitor anti-inflammatory drugs. It includes 2 large and 2 smaller trials that were originally designed to investigate the gastrointestinal safety of the 2 leading COX-2 inhibitors—celecoxib (Celebrex) and rofecoxib (Vioxx). The Celecoxib Long-Term Arthritis Safety Study (CLASS)1 compared celecoxib 800 mg/d to ibuprofen 2400 mg/d or diclofenac 150 mg/d in 7968 osteoarthritis or rheumatoid arthritis patients. Twenty-one percent of these patients were also concomitantly taking aspirin. The Vioxx Gastrointestinal Outcomes Research (VIGOR)2 study included 8076 rheumatoid arthritis patients randomized to rofecoxib 50 mg/d or naproxen 1000 mg/d. No aspirin therapy was allowed. The 2 smaller studies (designated studies 0853 and 0904) included 2020 osteoarthritis patients treated with rofecoxib 12.5 mg/d, nabumetone 1000 mg/d, or placebo.
Although the primary outcome measures for these studies were gastrointestinal, this analysis revealed an apparent increased risk of cardiovascular events associated with the use of COX-2 inhibitors. In the VIGOR study, 65 "serious" cardiovascular events (myocardial infarction, unstable angina, resuscitated cardiac arrest, sudden or unexplained death, ischemic stroke, or transient ischemic attack) were identified. There was a greater than 2-fold increase in the incidence of these events in the rofecoxib-treated group with a resultant risk ratio of 2.38 (95% CI, 1.39-4.00, P < 0.001). However, no such increased risk was seen with the use of rofecoxib in the 2 smaller studies (where only 12 total cardiovascular events were noted across all treatment groups), nor was any increased risk found with the use of celecoxib as compared to ibuprofen or diclofenac in the CLASS study. Overall incidence of myocardial infarction was similar for all treatment groups in the CLASS study (celecoxib 0.5%, ibuprofen 0.5%, diclofenac 0.3%), while the VIGOR study revealed a statistically significant disparity in myocardial incidence between the 2 agents (rofecoxib 0.4%, naproxen 0.1%).
Comment by John MacKnight, MD
Nonsteroidal anti-inflammatories (NSAIDs) are the most widely used drugs worldwide and are an invaluable adjunct to the management of pain and inflammation in a variety of conditions. The introduction of NSAIDs with selective COX-2 inhibition has led to intense interest in their use because of their favorable gastrointestinal toxicity profile. However, the data cited above have created an understandable concern in the medical community with regard to COX-2 safety. An analysis of the cardiovascular data from these trials is open to several interpretations.
First, these studies were not designed to prospectively assess cardiovascular risk. As such, it becomes difficult to feel confident that the data are as accurate and complete as if the study had been designed specifically to identify major cardiovascular events. In addition, direct comparison of the patient populations in the VIGOR and CLASS studies is problematic because of the heterogeneity of each study group. This is particularly pertinent in the VIGOR study because rheumatoid arthritis patients, who have a higher incidence of cardiovascular disease at baseline than osteoarthritis patients, comprised the entire study population.
Second, COX-2 agents have no discernible anti-platelet properties. VIGOR compared rofecoxib to naproxen, which is well known to cause significant and prolonged inhibition of platelet aggregation. Thus, naproxen may exhibit cardioprotective effects (similar to aspirin) not seen with the use of other NSAIDs. This certainly may have contributed to the different rates of cardiovascular events seen between the CLASS and VIGOR trials. Concomitant aspirin use by patients in the CLASS study may have further magnified this difference.
Finally, it is possible that COX-2 agents are indeed prothrombotic. COX-2 drugs do preferentially inhibit the production of platelet inhibitory prostacyclin but do not inhibit plaletet-activating thromboxane A2. This may create an imbalance in platelet activity and could lead to an increased rate of thrombotic events.
Physicians must base their treatment decisions on the best available evidence.3 A high-quality prospective study of potential cardiovascular risks with the use of COX-2 inhibitors is clearly indicated. In the interim, the low number of serious events relative to the number of patients studied should provide a measure of reassurance that no significant change in the use pattern of COX-2 agents is necessary at this point, pending additional prospective data. In addition, sufficient questions have been raised about the validity of the present data and its interpretations and the medical community should avoid labeling this valuable class of medications as clearly harmful. For those patients who are at high risk for cardiovascular events, the concomitant use of low-dose aspirin is a reasonable measure with the understanding that the protective effects of the COX-2 agents on gastrointestinal toxicity may be compromised.4 Until prospective data can clearly show otherwise, the use of COX-2 agents in ever-increasing numbers of patients would appear to outweigh their potential, if any, for increased cardiovascular risk.
Dr. MacKnight is Assistant Professor, Clinical Internal Medicine & Orthopaedic Surgery, Primary Care Team Physician, University of Virginia, Charlottesville, Va.
References
1. Silverstein FE, et al. JAMA. 2000;284(10):1247-1255.
2. Bombardier C, et al. N Engl J Med. 2000;343(21):1520-1528.
3. Food and Drug Administration. Cardiovascular safety review. Rockville, Md: Food and Drug Administration; 2001. Available at: www.fda.gov/ohrms/dockets/ac/01/briefing3677b2_06_cardio.pdf. Accessed December 3, 2001.
4. Steering Committee of the physicians’ research study group. N Engl J Med. 1989;321:129-135.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content