FDA warns Merck over its promotion of rofecoxib

BMJ 2001; 323 doi: (Published 06 October 2001) Cite this as: BMJ 2001;323:767
  1. Deborah Josefson
  1. San Francisco

    The US Food and Drug Administration has issued a warning letter to the pharmaceutical company Merck for misrepresenting the safety of their blockbuster anti-inflammatory drug, rofecoxib (Vioxx).

    Rofecoxib, a selective cyclo-oxygenase-2 (COX 2) inhibitor, was approved by the administration in May 1999 for the relief of acute pain, osteoarthritis, and dysmenorrhoea. The drug was heavily promoted by Merck and touted as safer than and superior to non-selective, non-steroidal anti-inflammatory drugs such as ibuprofen and naproxen.

    In its letter the FDA criticised Merck for playing down the possible risk of stroke associated with rofecoxib and for minimising potential drug interactions of rofecoxib with warfarin.

    The risk of stroke was found in an analysis of a large study, dubbed the VIGOR (Vioxx gastrointestinal outcomes research) trial, which compared 50 mg a day of rofecoxib with 500 mg twice a day of naproxen in patients with rheumatoid arthritis (New England Journal of Medicine 2000;343:1520-8).

    The VIGOR trial was a randomised, double blind, placebo controlled study of 4047 patients taking rofecoxib and 4029 taking naproxen. Patients were followed for an average of nine months.

    The study found that rofecoxib was significantly less ulcerogenic than naproxen and that patients taking rofecoxib had a 60% lower risk of serious gastrointestinal events—such as perforations, obstructions, and upper gastrointestinal bleeds—than patients taking naproxen. The relative risk was 0.4 (95% confidence interval, 0.2 to 0.8; P=0.005).

    The annual rate of these events was 1.4% among patients taking naproxen, compared with 0.6 % among patients taking rofecoxib.

    However, an analysis of the VIGOR study by cardiologist Eric Topol and colleagues at the Cleveland Clinic in Ohio (JAMA 2001;286:954-9) showed that patients taking rofecoxib had a higher relative risk of developing adverse cardiovascular events such as ischaemic strokes, unstable angina, and myocardial infarctions than the patients taking naproxen (relative risk 2.38 (95% confidence interval 1.39 to 4.00; P=0.002).

    Their study suggested that rofecoxib might be prothrombotic and urged further research to see if this was so. The FDA charged that Merck was aware of the cardiovascular risk associated with rofecoxib and minimised it in a press release and in its promotional materials.

    In a press release Merck responded that the VIGOR study's data falsely inflated the cardiovascular risk of rofecoxib because it compared the drug with naproxen, which has blood thinning properties similar to aspirin.

    View Abstract

    Sign in

    Log in through your institution

    Free trial

    Register for a free trial to to receive unlimited access to all content on for 14 days.
    Sign up for a free trial