November 29, 2009

Retraction—Combination treatment of angiotensin-II receptor blocker and angiotensin-converting-enzyme inhibitor in non-diabetic renal disease (COOPERATE): a randomised controlled trial

The COOPERATE study had previously been looked upon as providing support for combination ACE inhibitor and ARB therapy in proteinuric patients with non-diabetic kidney disease. After an initial letter of concern, the Lancet has retracted the paper after the results of an academic investigation indicated serious concerns surrounding this publication.  These include that the trial had not been approved by the hospital ethics committee, proper consent from patients had not been obtained, the involvement of a statistician could not be verified, the trial was not a double-blind study, and based on a sample of the medical records the authenticity of the data could not be proved.

Related Articles:

Chronic Kidney Disease, RAS Blockade
  • J Smithers

    In response to Dr. Lange’s post above, I think it is quite simple to look back with 20/20 hindsight at a paper that has now been retracted and state that we should have been suspicious all along.

    The reality in clinical nephrology is that we are often left with single centre studies that seem, based on the data presented, to be reasonably done studies. Though more or larger studies would be preferred, we often don’t get these in Nephrology.

    Clinicians are faced with the dilemma of what to do in the absence of robust data. Most of us will choose to take an action that seems reasonable based on the data available at the time, rather than do nothing and potentially undertreat large numbers of patients while waiting for a definitive trial that may never come.

    In the example of the COOPERATE study, I adopted combination therapy in IgA Nephropathy (and other non-diabetic proteinuric renal diseases) not because of anything that the editorial board of the Lancet said.

    I did it because I assumed the paper wasn’t fraudulent and that based on the data presented, this seemed to be better therapy than offered currently with little evidence of harm.

    I might argue that editorial boards need to demand access to the primary data so that it can be reviewed or, more likely, randomly audited.

  • DB Lange

    I think (hope) that most clinicians took the cooperate trial with a grain of salt. It was a small regional study looking at a select group of patients, that obviously limited external validity. That serious methodological flaws related to study design and setting also came to light, should come as no surprise. It is important that clinicians critically read and discuss the relevence of trials that suggest significant practice change, with colleagues and not just rely on a medical journal’s editorial board.

  • C Wang

    Productivity and quality control have to be balanced. But it’s also hard to implement a fair and non-biased system to monitor each study without affect the efficiency significantly.
    Current manuscript review system can be improved by blinding authors’ names to reviewers in order to receive fair evaluation at the beginning.