New Vioxx Study Projects Cases Of Heart Attacks
By ANNA WILDE MATHEWS
Staff Reporter of THE WALL STREET JOURNAL
October 6, 2004; Page A2
WASHINGTON -- A study led by a Food and Drug Administration safety official projects that the widespread use of Vioxx may have led to more than 27,000 heart attacks and sudden cardiac deaths before the drug's abrupt withdrawal last week by Merck & Co.
The number is in comparison to how many similar incidents would have occurred had the same patients been taking Celebrex, the Pfizer Inc. drug that competed with Merck's blockbuster arthritis treatment. The analysis specifically found that from Vioxx's approval in 1999, through 2003, an estimated 27,785 heart attacks and sudden cardiac deaths "would have been avoided" had Celebrex been used instead of Vioxx.
Those figures don't come from actual counts, but are projections based on findings from an analysis of a database of patients of Kaiser Permanente, the big health-maintenance organization.
Some of the paper's conclusions emerged in August. But the complete version, dated Sept. 30, goes farther on the possible impact of the painkiller, which was taken by approximately 20 million Americans before being withdrawn from the market last week.
---
A TALE OF TWO DRUGS
A study led by an FDA researcher projects thousands of serious heart problems would have been avoided had people taken Celebrex not Vioxx if people had taken Celebrex not Vioxx between 1999 and 2003
DAILY VIOXX DOSE - NUMBER OF PRESCRIPTIONS - Heart attacks, Sudden cardiac deaths that wouldn't have occurred
Less than 25 mg 76,406,000 14,845
More than 25 mg 16,385,000 12,940
Total 92,791,000 27,785
Source: internal FDA documents
---
In providing its potential estimate of the adverse impact, the study is a new headache for Merck, which is facing a flood of lawsuits that will question whether the company should have taken stronger action sooner in response to concerns about Vioxx. Merck quickly withdrew the drug after a clinical trial found that people taking a low dose of Vioxx for more than 18 months were twice as likely to have a heart attack or stroke as those taking a placebo. Earlier research had suggested a possible link between the drug and cardiovascular problems.
A Merck spokesman said in a statement that the company "cannot comment on the full study, as we have not yet had the opportunity to review it." He added that Merck believed that controlled clinical trials are "the best way to evaluate the safety of medicines," while "epidemiology studies are limited in their ability to understand effects" because researchers can't fully control for differences between different groups of patients.
The study led by the FDA researcher hasn't been released to the public, but a copy has been requested by Senate Finance Committee Chairman Charles Grassley, an Iowa Republican, who is investigating how the FDA handles safety concerns.
Its lead author, David Graham, is associate director for science in the FDA drug center's office of drug safety. Other authors are from Kaiser and Vanderbilt University. The study was initiated by the FDA and funded by an agency contract with Kaiser, which contributed some support.
Steven Galson, the acting director of the FDA's drug center, said he first saw the study yesterday and other officials got it last Thursday, the day Merck announced the withdrawal, so it is "too early to say" whether the agency agrees with the findings of the authors.
"It would be irresponsible of us to react that quickly," he said. The study also hasn't yet been peer-reviewed by outside scholars. He said the projection of the heart problems tied to Vioxx "relies on a lot of assumptions," and such conclusions are "speculative," but the report appeared "carefully done."
The study drew on data from about 1.4 million Kaiser patients, who had taken one of the painkillers called nonsteroidal anti-inflammatory drugs, or NSAIDs. That included 40,405 patients who had taken Celebrex and 26,748 who had taken Vioxx.
The projections of the potential impact came from figuring out a rate of increased risk among the Kaiser patients, and then applying it to the number of American prescriptions for Vioxx in the years 1999 through 2003. Of the additional 27,785 serious heart problems the study projected for the entire U.S., 12,940 would have come in patients taking high doses of Vioxx, and 14,845 with low doses.
The actual numbers in the study database were far smaller -- the study estimated that of 58 cases of heart attacks or sudden cardiac death among patients in the database who were taking the low dose of Vioxx, 21 wouldn't have happened had they all been taking Celebrex. For the high dose, it was 9.7 serious heart problems out of 10 that wouldn't have happened.
The study found that people taking a high dose of Vioxx were 3.69 times as likely to have a serious cardiac event as people taking Celebrex, while the ratio for people taking a low dose of Vioxx was 1.5. It concluded that there was no evidence of a "substantial benefit with the high-dose strength" of Vioxx that would "counter-balance the level of cardiovascular risk" shown in various studies of the drug.
Write to Anna Wilde Mathews at anna.mathews@wsj.com |
