More Evidence on Cardiovascular Risks of COX-2 Inhibitors, Other NSAIDs

Laurie Barclay, MD

January 24, 2005

Jan. 24, 2005 — Four studies and an editorial in the Jan. 24 issue of the Archives of Internal Medicine shed light on the cardiovascular risks of selective cyclooxygenase-2 (COX-2) inhibitors and nonselective COX inhibitors, as well as on the risks of gastrointestinal (GI) bleeding when used in combination with warfarin.

"The COX-2 inhibitor story continues to evolve, even now following the withdrawal of rofecoxib from the worldwide market in October 2004," the editors write. "The four articles on COX-2 inhibitors in this issue of the Archives demonstrate continued reason for concern about adverse cardiovascular effects of both rofecoxib and celecoxib (though less so for the latter) as well as serious concern about the apparent overuse of these drugs in the prescription marketplace."

The editors also note that the Archives has published at least 10 clinical research articles on the cardiovascular and other effects of the COX-2 inhibitors during the past five years but that their adverse cardiovascular effects have been recognized only recently.

Risk of GI Bleed With Warfarin Similar for COX-2 Inhibitors and Nonselective NSAIDS

In the first study, Marisa Battistella, BScPhm, PharmD, and colleagues write, "Little is known about the risk of upper GI hemorrhage during the concomitant use of warfarin and selective COX-2 inhibitors. We examined the association between the concomitant use of warfarin and nonselective nonsteroidal anti-inflammatory drugs (NSAIDs) or selective COX-2 inhibitors in older adults hospitalized for upper GI hemorrhage." The authors are from the University Health Network of the University of Toronto in Ontario, Canada.

This nested case-control analysis conducted over one year in Ontario used multiple linked healthcare databases to identify a cohort of 98,821 patients older than 66 years and continuously prescribed warfarin. Using a control-case ratio of 4:1, the investigators compared prescription records prior to hospitalization in 361 case patients (0.3%; those admitted to the hospital with upper GI hemorrhage while taking warfarin) with those of age- and sex-matched controls who were also receiving warfarin.

After adjusting for other potential confounders, case patients were significantly more likely than control patients to also be taking nonselective NSAIDs (odds ratio [OR], 1.9; 95% confidence interval [CI], 1.4 - 3.7), celecoxib (OR, 1.7; 95% CI, 1.2 - 3.6), or rofecoxib (OR, 2.4; 95% CI, 1.7 - 3.6) before hospitalization.

Study limitations include inability to account for some potentially important factors such as smoking, alcohol consumption, and the use of nonprescription analgesics; higher prevalence of GI comorbidity in the case patients at presentation; low absolute number of events precluding reliable subgroup analyses; limited generalizability to younger patients or settings with different drug policies for longer durations of follow-up; and lack of data on international normalized ratios in any patients.

"Patients taking warfarin concomitantly with selective COX-2 inhibitors have an increased risk of hospitalization for upper GI hemorrhage," the authors write. "The risk appears similar to that of patients simultaneously taking warfarin and nonselective NSAIDs."

Some of the authors report various financial relationships with the Canadian Institutes of Health Research, the University of Toronto Drug Safety Research Group, and/or Pfizer, Inc.

Widespread Overuse of COX-2 Inhibitors May Be Related to Cardiovascular Events

A second study suggests that increased cardiovascular events attributable to COX-2 inhibitors might be related to widespread overuse, even by patients unlikely to benefit from this novel but expensive pharmaceutical class.

"We found a rapid nationwide shift away from older, inexpensive drugs with better established safety and efficacy to newer, costly drugs with no real history," author G. Caleb Alexander, MD, MS, from the MacLean Center for Clinical Medical Ethics at the University of Chicago in Illinois, says in a news release. "What we saw was widespread, rapid adoption of an interesting and promising but expensive and largely untested medication by millions of people with little or nothing to gain from long-term use."

The investigators analyzed data from the National Ambulatory Medical Care Survey (1999-2002) and the National Hospital Ambulatory Medical Care Survey (1999-2001), reflecting community and hospital-based outpatient practices nationwide. The primary outcome measure was the proportion of patient visits in which COX-2 inhibitors were prescribed, stratified by risk of adverse GI events from NSAIDs.

For the visits in which either a COX-2 inhibitor or NSAID was prescribed, the frequency of COX-2 inhibitor use increased from 35% in 1999, to 55% in 2000, to 61% in 2001 and 2002. Nearly two-thirds of this growth reflected increases in COX-2 inhibitor use among patients in whom NSAIDs could be used. For the subgroup of patients with the lowest risk for adverse events from NSAIDs, COX-2 inhibitor use increased from 12% in 1999 to 35% in 2002.

"We pursued our analysis because of concerns that this was an example of overuse of a newer, more expensive class of medications," says coauthor Randall S. Stafford, MD, PhD, from Stanford University in California. "As with other examples in the treatment of high blood pressure, diabetes and some infections, there appears to be a tendency for new drugs to be used in a wider array of situations than supported by scientific studies."

The MacLean Center for Clinical Medical Ethics at the University of Chicago, the National Institute of Aging, and the Agency for Healthcare Research and Quality supported this study.

Observational Study Finds No Increase in Cardiovascular Thrombotic Events

A third study, from the University of Maryland in Baltimore, was an observational cohort study looking at the cardiovascular risk of COX-2 inhibitors compared with nonspecific NSAIDs in Maryland Medicaid enrollees. This high-risk population was 70% female, 50% African American, and 30% older than 50 years.

Of the 6,250 patients, 1,005 patients were using COX-2 inhibitors and 5,245 patients were using a nonnaproxen NSAID. Overall, 12% of patients had one or more cardiovascular thrombotic events after treatment within the follow-up period, but the propensity-adjusted odds ratio showed no significant effect of COX-2 inhibitor use on this outcome. The investigators concluded that COX-2 inhibitors did not increase cardiovascular risk over nonnaproxen NSAIDs in this population.

No pharmaceutical company supported this study. Three authors report various financial arrangements with Pfizer, Inc., and/or Merck.

Systolic BP May be Increased With Rofecoxib But Not Celecoxib or Naproxen

The final study report was that of the Celecoxib Rofecoxib Efficacy and Safety in Comorbidities Evaluation Trial (CRESCENT). In this double-blind trial, patients with type 2 diabetes, hypertension, and osteoarthritis were randomized to treatment with 200 mg of celecoxib once daily (n = 136), 25 mg of rofecoxib once daily (n = 138), or 500 mg of naproxen twice daily (n = 130) for 12 weeks.

After six weeks of therapy, patients' mean 24-hour systolic blood pressure was increased significantly by rofecoxib (from 130.3 ± 1.2 mm Hg to 134.5 ± 1.4 mm Hg; P < .001) but not by celecoxib (from 132.0 ± 1.3 mm Hg to 131.9 ± 1.3 mm Hg; P = .54) or naproxen (from 133.7 ± 1.5 mm Hg to 133.0 ± 1.4 mm Hg; P = .74). Furthermore, the proportion of patients with controlled hypertension at baseline who developed ambulatory hypertension was significantly greater with rofecoxib (30%) than with celecoxib (16%; P = .05), but not significantly greater than with naproxen (19%).

Pharmacia Corp. and Pfizer, Inc., supported this study and have financial arrangements with several authors.

Questions Raised Concerning Drug Policy, Evidence

An accompanying editorial reviews the events and data leading to rofecoxib's withdrawal from the market on Sept. 30, 2004, as well as the roles of postmarketing surveillance and observational studies of drug safety.

"The dramatic withdrawal of such a widely used and widely promoted drug five years after it was introduced to the market raises many questions concerning drug policy, scientific evidence, and treatment alternatives," write Daniel H. Solomon, MD, MPH, and Jerry Avorn, MD, from Brigham and Women's Hospital in Boston, Massachusetts. "A more thoughtful approach to these issues by regulators, industry, clinicians, and researchers could help avoid similar episodes in the future."

They warn against assuming a class effect when evaluating a drug's safety, as exemplified by cerivastatin, troglitazone, and bromfenac, each of which was withdrawn from the market not long after approval because of liver toxicity or other adverse effects.

"Several lessons can be learned from the 5-year experience and eventual withdrawal of rofecoxib from the market. First, the current postmarketing surveillance system does not work. If the FDA is to continue to approve drugs rapidly, we should not expect that all safety issues will be understood prior to a drug's approval," Drs. Solomon and Avorn write. "The agency will need to be more effective in requiring specially designed premarketing clinical safety trials when phase II or small phase III trials suggest reason for specific concerns. Such a system should be rapidly responsive and objective, and safeguards to protect this agenda against industry influence must be put in place."

Drs. Solomon and Avorn have received research grant support from Merck and Pfizer, manufacturers of COX-2 inhibitors.

Arch Intern Med. 2005;165:158-160, 161-168, 171-177, 181-186, 189-192

Reviewed by Gary D. Vogin, MD


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.