Suspect Drug Research Blamed for Massive Death Toll
Research misconduct can ruin everything from scientific careers to institutional reputations and public confidence in science. But in a paper published 2 weeks ago, two British cardiologists claimed that mis conduct in their field may have had a far greater toll. Tainted research by Don Poldermans, a disgraced cardiologist who was at Erasmus MC in the Netherlands, may have led to the deaths of 800,000 people in Europe, Darrel Francis and Graham Cole of Imperial College London wrote in a provocative article that appeared briefly in the European Heart Journal (EHJ) and was then withdrawn.
Poldermans, a prominent researcher who published more than 300 papers, was fired in November 2011 after a university investigation concluded that he had engaged in misconduct, including data fabrication. He was the lead author on two influential trials examining whether β-blocker drugs can protect patients undergoing surgery that doesn't directly involve the heart; those studies helped shape guidelines adopted in 2009 by the European Society of Cardiology (ESC) that recommended using the drugs. (U.S. guidelines are more cautious.) When Poldermans's studies are omitted, Francis and Cole say, the evidence shows that the recommendations don't save lives but endanger them.
The accusatory paper was removed from the EHJ's website less than 48 hours after it appeared. It hadn't undergone peer review, as it should have, says Thomas Lüscher of the University of Zurich in Switzerland, the journal's editor; an official retraction was posted on 23 January, and the paper is now under review. But Cole and Francis say the staggering number of deaths they calculated was based on published data, and their claim has reignited a debate about giving β blockers to patients about to undergo surgery that might stress the heart. It is also a reminder, some scientists say, of the huge effects that a few uncertain and potentially flawed studies can have on clinical practice. "This is unfortunately what happens when you write a guideline that affects large numbers of people in a relatively common situation," Francis says.
Defenders of the guidelines counter that the estimate of 800,000 deaths is wildly inflated. It disregards explicit cautions in the guidelines, Lüscher says. Poldermans says he has seen the paper but he declined to comment; he is waiting for the review and revision of the paper, which Francis and Cole hope will be republished shortly.
http://www.sciencemag.org/content/343/6170/473.full