Key Mediterranean Diet Trial Retracted

— But republished results show same advantage after excluding protocol violations

Last Updated July 11, 2018
MedpageToday

The landmark PREDIMED randomized trial of a Mediterranean diet has been influential and highly cited since its 2013 publication in the New England Journal of Medicine. But now it has been retracted and republished after hidden protocol changes and randomization violations came to light.

The tip-off came from an analysis of randomized trials looking for improbable distributions of baseline data, with PREDIMED flagged for variables inconsistent with randomization, according to an editor's note appearing with the retraction.

"There were departures from the randomization procedures that had been specified in the protocol that were not described in our original report," the researchers explained after reanalysis and republication. Of the trial's 7,447 Spanish participants, 425 participants shared a household with a previously enrolled participant and got the same diet assignment without randomization.

"After the study had begun, the steering committee approved this protocol change. The protocol was not amended, and this protocol change was not described in the original report published in the Journal," they noted.

Another problem uncovered in the reanalysis was that one of the study sites assigned participants by clinic rather than individually, which affected another 467 participants. A second study site only inconsistently used the randomization tables.

However, reanalysis excluding patients from both those sites and the non-randomized second household members didn't change the findings on the primary endpoint of myocardial infarction, stroke, or death from cardiovascular causes.

The Mediterranean diet still came out significantly superior to the control diet (advice to reduce dietary fat) with a hazard ratio of 0.71 when supplemented with extra virgin olive oil and 0.68 when supplemented with nuts, compared with hazard ratios in the original analysis of 0.70 for both.

Eleven other trials were reanalyzed after being flagged and corrections were issued on five that appeared in NEJM at the same time:

  • High-Dose Atorvastatin after Stroke or Transient Ischemic Attack
  • Treatment of Periodontitis and Endothelial Function
  • Extended Antiretroviral Prophylaxis to Reduce Breast-Milk HIV-1 Transmission
  • Effect of Bronchoconstriction on Airway Remodeling in Asthma
  • Horse versus Rabbit Antithymocyte Globulin in Acquired Aplastic Anemia

In each, "standard errors were inadvertently reported as standard deviations, or vice versa," the editor's note said.

Primary Source

New England Journal of Medicine

Source Reference: Estruch R, et al "Primary prevention of cardiovascular disease with a Mediterranean diet supplemented with extra-virgin olive oil or nuts" N Engl J Med 2018; DOI:10.1056/NEJMoa1800389.