Use of β blockers during surgery may have caused 10 000 deaths in the UK

BMJ 2013; 347 doi: (Published 05 August 2013) Cite this as: BMJ 2013;347:f4914
  1. Nigel Hawkes
  1. 1London

Up to 10 000 deaths may have been caused in the United Kingdom by following a guideline that can no longer be relied upon, says an analysis by a group from Imperial College in London.

Writing in Heart online, the group said that a guideline recommending the use of β blockers to protect the heart during non-cardiac surgery relied heavily on results from a trial led by a now discredited Dutch researcher, Don Poldermans.1 To make matters worse, Poldermans also chaired the committee that wrote the guideline.

Darrel Francis and colleagues said that there was no proof that the DECREASE trial (Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography) that Poldermans led was dishonest, but data were stored haphazardly and where they could be located, the Dutch investigation into Poldermans found them to be fictitious.

The DECREASE family of studies “fell far short of the standards assumed by clinical readers,” the group said.

As a result, the evidence for the use of β blockade needed to be reassessed with the DECREASE results excluded. When the group did this, they found that the remaining evidence indicated that β blockade led to a 27% increase in deaths. While the analysis did show fewer non-fatal heart attacks, the increase in mortality had “overwhelming clinical significance.”

The guideline is applicable to around 2.5 million high or medium risk operations in the UK every year, from which there are 47 286 deaths within 30 days.

The group concluded, “Refraining from this European Society of Cardiology guideline would therefore be expected to prevent up to 10 000 iatrogenic deaths each year in the UK.”

In reality it is not known how universally the guideline is followed, so the number may be smaller.

The scandal over Poldermans emerged in 2011, but the guideline remains unchanged two years later. The chairman of the ESC Clinical Practice Guidelines Committee, Jose Louis Zamorano, told that the society was taking the new analysis “very seriously” and would be convening a task force to decide urgently if further action was required.2

The Imperial College team said that guideline bodies should retract recommendations based on fictitious data “without further delay,” and should not waste time disputing over who was to blame. As for those who were unconvinced, “they might best channel their energy into a further randomised trial which should be designed carefully and conducted honestly,” the group added.

They said that future guidelines should be accompanied by a commitment from named individuals to retract them immediately if the advice given was later found to be harmful.

While the original guideline was dominated by DECREASE, the new analysis is also dominated by a single large trial, POISE (Perioperative Ischemic Evaluation study). This is appropriate, said the team, because the POISE trial was by far the largest carried out and was well conducted. “Without it there is little remaining evidence base,” the group said.


Cite this as: BMJ 2013;347:f4914