The debate over the results of EXCEL shows no sign of abating, and there seems to be a division between cardiac surgeons and interventional cardiologists about which side of the argument they fall on, with surgeons generally supporting EACTS, and interventional cardiologists generally supporting the EXCEL leadership. Although some interventional cardiologists do not believe the disagreements will change their practice, they raise concerns that their working relationships with surgeons may suffer.
The decision in December 2019 by the European Association of Cardio-Thoracic Surgeons (EACTS) to withdraw support for the left main section of its myocardial revascularisation guidelines has reverberated around the cardiology community; many societies have since called for an independent review of EXCEL data which were used to inform the guidelines. But leading interventional cardiologists say that despite the furore, as yet, nothing has changed.
Nick Curzen (University Hospital Southampton NHS Foundation Trust, Southampton, UK), says multiple other factors inform his decision-making. “I will offer the patient an informed choice, laying out the pros and cons of each. For percutaneous coronary intervention (PCI) these are: a quick admission, and a quick recovery, with less trauma, but there is a higher revascularisation rate; therefore, there is less chance of needing to come back after coronary artery bypass graft (CABG). There are no differences in death rates, and probably none in MI [myocardial infarction], between the procedures.”
He says that neither of the two main areas of contention raised by the Newsnight programme, and subsequently acted upon by EACTS, “would affect my interpretation of the clinical value of the trial or its contribution to the evidence base”. He points out that the difference in all-cause mortality between CABG and PCI “is a secondary endpoint and, while interesting, does not set off alarm bells for me about PCI”.
David Hildick-Smith (Brighton and Sussex University Hospitals NHS Trust, Brighton, UK) agrees that he too will treat patients “the same as before”. He tells Cardiovascular News that there are “a wealth of good trials—EXCEL, NOBLE, SYNTAX, PRECOMBAT etc” on which to base decisions about therapy, giving a consistent message” that patients are more likely to require repeat intervention after stents than after surgery, “but to get the benefits of surgery you have to have the operation and not all patients are keen on this. In the end, it is the patient who decides what treatment they will submit to, not the doctor.”
“A very public disagreement”
Both Curzen and Hildick-Smith have expressed concern about how some of the parties in the debate have handled the situation. Hildick-Smith declares himself “disappointed” with the EACTS’ leadership, accusing them of “inflammatory and sensationalist use of relative risk in a way that is likely to be misunderstood by the general public; the tabloid language used has not helped relations”.
He adds: “It is a shame that the surgical leadership of EACTS has chosen to make this a very public disagreement. Some fairly egregious accusations have been made and I think the leadership of EXCEL has behaved well in the face of the provocation.”
Similarly, Curzen is critical of the handling of the situation “by some of the protagonists and by the BBC Newsnight team”. He thinks “the lack of balance” will undermine patients’ confidence in PCI, which he says is “most regrettable” and could have been avoided. Furthermore, Curzen is surprised at EACTS’ move to withdraw support from the guidelines—“allowing for the fact that I have not seen all of the BBC leaked data”. “The British Cardiovascular Interventional Society (BCIS) Council has not felt that there are grounds to do so,” he points out.
Adrian Banning (John Radcliffe Hospital, Oxford, UK) also describes events as “regrettable”, with “a lack of transparency about some of the things people have said, and this has resulted in confusion. Ultimately, not only does this undermine clinicians’ confidence, but it also threatens the confidence patients have in clinical research.”
Independent review of the data
However, cardiothoracic surgeon David Taggart (University of Oxford, John Radcliffe Hospital, Oxford, UK), an EXCEL investigator who gave a heavily critical talk about the trial at EACTS 2019, and who appeared in the Newsnight programme, is calling for an independent review. He says: “How physicians are now going to approach these patients is at the heart of what the fundamental problem is. Concerns about the trial now go far beyond EACTS’ decision to withdraw support.” He believes that the review, conducted in a “fully independent way”, is necessary to restore faith in the guidelines.
“Basically, I want transparency. This is not a sterile academic debate—it puts individual doctors in an invidious position. To properly restore faith and confidence in the guidelines, we need a fully independent reanalysis by a group of clinician scientists who were not involved in either drawing up the guidelines or in the trial itself.” Taggart, because of his concerns about the findings of EXCEL, withdrew his name from the New England Journal of Medicine (NEJM) paper in which the five-year findings were published.
Banning says that, although an independent review is a good idea: “I doubt very much it will change the key messages from the trial, or the summation of evidence used to produce the guidelines. I worry we will have expended a lot of effort to end up broadly where we started.”
Much of the debate has played out on Twitter and, generally, the opinions being expressed have divided along professional lines. Banning judges Twitter’s impact to have been negative: “I do not think it has helped at all as it allows rumour and ‘Chinese whispers’ to propagate. I am in favour of social media, but this highlights a potential disadvantage. Institutions and societies cannot react fast enough to Twitter storms, and then the Twitter focus moves on.”
Others, however, view it more positively. Taggart does not use social media himself, but believes Twitter could have helped in the longer-term: “Although the controversy over EXCEL is potentially, at least superficially, damaging to doctors, it also means that there will be no other trial whose conduct is not subject to vigorous scrutiny in the future. The ultimate result could be better trials.” But on the flip side, he adds, “we can only do trials if patients are willing to participate, and anything that shakes their confidence is detrimental.”
Curzen thinks Twitter activity “has been interesting. The value for me was the extra wealth of data offered by colleagues on Twitter in trying to support their arguments, some of which I had not seen before. That is Twitter at its best.”
For Hildick-Smith: “So long as people remain respectful and avoid trolling, I think it is fair enough to see commentary on social media. Some incisive early commentary was seen on EXCEL.”
A broken heart team?
As Twitter underlined, opinion about EXCEL appears to be splitting along specialty lines, and could have the potential to damage working relationships within heart teams, as well as the trust enjoyed between different professions. All of those that Cardiovascular News spoke to emphasise the importance of strong interdisciplinary relationships. Curzen stresses “the value of discussion between the disciplines, focused on the patient and not the doctors, is extremely effective, and demands respect and cooperation between surgeons and interventionalists.”
Banning agrees: “If we lose the heart team ethic, patients will lose out.” He fears that it could signal a return to the “dark days of the early drug-eluting stents era—before the SYNTAX trial where there was a lack of cohesive teamwork”, and calls for an end to tribal allegiances. Curzen is also aware that “some bad feeling” has been generated, and Taggart acknowledges: “There has been a split between surgeons and interventional cardiologists”. He was optimistic that it would be a “temporary schism”. “EACTS and the cardiothoracic societies want to work with cardiologists,” he says.
Hildick-Smith declares that a “period of reflection by the surgeons would be sensible. A meta-analysis of NOBLE and EXCEL, and perhaps also FREEDOM and PRECOMBAT, will demonstrate that surgery is a very good option for those willing to have surgery. For other patients there are stents, and results from stent implantation are very good too.”
It is time to move on, says Curzen: “I have received all that I need from EXCEL and NOBLE now. I suspect, however, that the EXCEL investigators are now obliged to publish a supplementary paper relating to the periprocedural enzyme data.”