Just two weeks after Merck & Co. pulled its cyclooxygenase-2 (COX-2) inhibitor rofecoxib (Vioxx) from the market, Pfizer Inc. told doctors that its COX-2 inhibitor valdecoxib (Bextra) increases the risk of heart attack or stroke in patients undergoing coronary artery bypass surgery.

In a letter to healthcare professionals, Pfizer said that in two trials in which valdecoxib was used in the setting of heart surgery, there was a greater appearance of cardiovascular events.

The trials utilized valdecoxib and parecoxib -- an investigative intravenous formulation -- in patients undergoing open heart surgery. The first study was published last year; the second study was just completed, the Pfizer statement said. Pfizer noted that valdecoxib is not approved in any surgical setting in the United States.

"This is not the same story as rofecoxib, nor do these results compare with the way Bextra is being prescribed today," commented Lee Simon, MD, associate professor of medicine at Harvard Medical School, Boston, Mass. "This is a very different dose, a different patient population, and a different use of the drug."

Dr. Simon, who discussed the results of the surgery trial with Pfizer officials during the 2004 annual scientific meeting of the American College of Rheumatology in San Antonio, Texas, said the dose of valdecoxib used in the trial (40 mg) was double the 10 mg to 20 mg normally prescribed for arthritis and other pain.

Pfizer said in the statement that valdecoxib [Bextra] is indicated for the relief of signs and symptoms of osteoarthritis and adult rheumatoid arthritis and for the treatment of primary dysmenorrhea. The medication was approved and introduced in the U.S. market in 2001. Available clinical information for valdecoxib suggests there is no increased risk of cardiovascular thromboembolic events in people treated for osteoarthritis and rheumatoid arthritis.

"I still take Bextra myself," Vibeke Strand, MD, adjunct clinical professor of medicine in the division of rheumatology/immunology at Stanford University Medical Center, Stanford, Calif, said at the ACR meeting. "I don't think the drug should be removed from the market."

"There were several problems with the CABG [coronary artery bypass graft] study," Dr. Strand continued. "The patients who were enrolled in the study were at very high risk, making it difficult to sort out whether the events seen were due to valdecoxib or to the underlying condition of the patients. However, there does seem to be a signal with the higher doses, which might indicate a problem with valdecoxib. We saw a similar dose-response increase of events with rofecoxib as well."

Dr. Simon said that on the basis of the new information, he would still prescribe the medication for his patients -- but only after fulfilling a number of criteria, including determination that the patient actually needed a COX-2 inhibitor and could not have the pain controlled with other medications, and if the patients were unable to obtain relief with celecoxib (Celebrex), which has yet to be implicated in heart disease.

He said he was concerned that while published studies have examined celecoxib in 31000 patients and rofecoxib in 28000 patients, valdecoxib experience has only been reported in about 9700 patients.

"I certainly believe that the COX-2s are not all associated with a cardiovascular signal," Dr. Strand said. "I think there is one for etoricoxib,one for rofecoxib,and now we have seen one for valdecoxib. But celecoxib to date does not appear to have one. We as physicians need to look at these COX-2s as individual drugs with variability, just as the non-steroidals are variable."

The US Food and Drug Administration is expected to convene a panel in January to review issues raised by the new studies with Bextra and Vioxx and perhaps other COX-2 drugs now available and in development.

"I think that Bextra will remain on the market, but there will be a new warning box with the prescribing information," said Edward Keystone, MD, professor of medicine at the Arthritis Center of Excellence at University of Toronto, Canada, who responded to the Pfizer announcement while atttending the annual ACR meeting. "I would not prescribe Bextra to any of my patients with heart disease. There could be a problem in generally prescribing the drug, in that many people who have heart attacks have no symptoms of the disease and also may not have common risk factors either."

In the Pfizer statement, Joseph Feczko, MD, president of worldwide development for Pfizer, said, "Bextra is an important treatment option for patients faced with debilitating and chronic pain associated with osteoarthritis and rheumatoid arthritis. At the same time, as is true with any medicine, all benefits and risks need to be considered by physicians when treating their patients. This communication is intended to reinforce our commitment to share information with physicians about our product."

Commenting on the release of the letter from Pfizer to all health-care providers, Dr. Strand said, "I think Pfizer is doing the right thing; we need to further study this drug."