December 9, 2019
EACTS Withdraws Support of Left Main Chapter of European Guidelines for Myocardial Revascularization After BBC Investigation of the EXCEL Trial
December 9, 2019—Professor Domenico Pagano, MD, Secretary General of the European Association for Cardiothoracic Surgery (EACTS), issued a statement in response to the report of the United Kingdom’s BBC Newsnight investigation on the EXCEL trial and the society’s subsequent withdrawal of support for the current guidelines related to myocardial revascularization for left main coronary artery disease (LMCAD) that were established jointly by EACTS and the European Society of Cardiology (ESC).
EACTS noted that new trial data help inform the clinical guidelines used to advise patients on the best available treatment options, and the reported outcomes of the EXCEL trial were included with those used to develop the joint 2018 EACTS/ESC Clinical Guidelines for Myocardial Revascularization, made available online in August 2018 and published by Franz-Josef Neumann, MD, et al in the European Heart Journal (2019;40:87–165).
In contrast to EACTS, the ESC rejected the BBC report's claim that the guidelines may have caused harm to patients and stands by the guidelines, which ESC says were based on more than the EXCEL trial. Additionally, the Society for Cardiovascular Angiography and Interventions (SCAI) countered the findings of the BBC report and disagreed with the EACTS' action. The SCAI continues to endorse the European and United States treatment guidelines. The society's full statement is provided at the end of this article.
The BBC report (available here) called into question the methods and conclusions of the EXCEL trial, which compared coronary artery bypass graft (CABG) surgery with percutaneous coronary interventions (PCIs) in patients with LMCAD. The EXCEL investigators concluded that there was no significant difference in patient outcomes, regardless of which treatment option was used.
A key issue is that the definition of myocardial infarction (MI) used in the EXCEL trial differed from the Universal Definition of Myocardial Infarction ("Universal Definition"). In 2012, the Third Universal Definition was published by Professor Jeroen J. Bax, MD, et al in Journal of the American College of Cardiology (2012;60:1581–1598). This definition was revised and updated in October 2018 by Professor Kristian Thygesen, MD, et al in Journal of the American College of Cardiology (2018;72:2231–2264). The definition reflects a joint consensus of the ESC, American College of Cardiology, American Heart Association, and World Heart Federation.
SCAI advised that the EXCEL trial used the SCAI definition of MI, which does not align with the Third Universal Definition, which was active during the period of the study. The SCAI statement (presented in full below) further explains the differences of the two definitions and the reason for the use of the SCAI definition in the EXCEL trial. The SCAI's expert consensus document on the definition was published in 2013 by Issam D. Moussa, MD, et al in Journal of the Amercan College of Cardiology (2013;62:1563–1570).
Neither the EACTS statement nor the BBC report specified that the SCAI definition was used in the EXCEL trial.
As previously reported in Cardiac Interventions Today, EXCEL recruited 2,905 patients with LMCAD at 126 sites in 17 countries between September 2010 and March 2014. Eligible patients (n = 1,905) with LMCAD and site-assessed low or intermediate CAD complexity (SYNTAX score ≤ 32) were randomized to revascularization with fluoropolymer-based, cobalt-chromium, everolimus-eluting stents (EESs; n = 948) or CABG (n = 957). The EXCEL trial was funded by Abbott Vascular and used the company’s Xience Prime EES, Xience V EES, Xience Xpedition EES, or Xience Pro EES (used only outside the United States) to compare PCI and CABG.
According to the BBC report, there is debate about whether the Universal Definition or the SCAI definition is a better measure of MI; the EXCEL investigators continue to stand by their choice of definition. The BBC reported that the EXCEL investigators had said that they would also publish data applying the Universal Definition but that this has not been done.
Newsnight reported it has seen these unpublished data and that they show that under the Universal Definition, patients in the EXCEL trial treated by PCI had 80% more MIs than those treated by CABG. Newsnight stated it had spoken to experts who support the credibility of the data; lead investigators from EXCEL told Newsnight they disputed the accuracy of this report.
The 3-year results from the EXCEL trial were presented at TCT 2016, the 28th annual Transcatheter Cardiovascular Therapeutics scientific symposium in Washington, DC, and published by Coprincipal Investigator Gregg W. Stone, MD, et al in The New England Journal of Medicine (2016;375:2223–2235). The article concluded, "In patients with left main coronary artery disease and low or intermediate SYNTAX scores by site assessment, PCI with everolimus-eluting stents was noninferior to CABG with respect to the rate of the composite endpoint of death, stroke, or myocardial infarction at 3 years."
Additionally, the BBC reported that as the guidelines were being drafted, the EXCEL trial's data safety monitoring board raised concerns about the higher mortality rate among patients who received stents.
In September 2019, the 5-year results from the EXCEL study were presented by Dr. Stone at TCT 2019, the 31st annual Transcatheter Cardiovascular Therapeutics scientific symposium in San Francisco, California. The study was published simultaneously by Dr. Stone et al in The New England Journal of Medicine (2019;381:1820–1830). The EXCEL investigators reported that all-cause death occurred in 13% of patients in the PCI group and 9.9% of patients in the CABG group (difference, 3.1%; 95% confidence interval, 0.2%–6.1%). They observed that of the 30 excess deaths in the PCI arm, 18 were adjudicated as noncardiovascular deaths, five as definite cardiovascular deaths, and seven as undetermined cause.
The BBC report cited a complaint from a former EXCEL investigator that these elevated mortality-rate data were not sufficiently highlighted.
- The following is Professor Pagano’s statement issued by EACTS:
“We welcome Newsnight’s scrutiny of the EXCEL trial and the findings. These show patients with left main coronary artery disease treated with stents are 35% more likely to die than those treated with conventional open-heart surgery.
“The Council of the EACTS has considered the analysis of the data that BBC Newsnight has shown us. It is a matter of serious concern to us that some results in the EXCEL trial appear to have been concealed and that some patients may therefore have received the wrong clinical advice.
“Following the information presented to us by Newsnight, the EACTS Council has unanimously decided, with immediate effect, to withdraw our support for the Left Main Chapter of the joint 2018 EACTS-ESC Clinical Guidelines for Myocardial Revascularization. If the information on the trial is proven to be correct, the recommendation is unsafe. On behalf of the EACTS Council, I have written to the ESC to invite them to work with us to develop a new joint section of the guidelines as a matter of urgency.
“We deeply regret the obvious concern that this will raise for some patients and their families. We recommend that patients seek the advice of the multidisciplinary heart team at their hospital before deciding which treatment option is most appropriate for them.”
The EACTS announcement further stated, “We recognize that if the data and the analysis Newsnight has carried out are correct, as they appear to us to be, patients have been subjected to an increased risk of death. That’s why the EACTS Council voted unanimously to withdraw our support for the guidelines on left main disease with immediate effect. We urge our members to disregard the guidelines relating to left main disease for the time being.”
EACTS noted that at the annual conference this year, leading international experts were invited to debate the findings from the latest clinical trials, including EXCEL. The society live-streamed the 4-hour discussion and provided the link to it in its announcement. The relevant portion begins at the 2-hour mark here. Finally, the society advised that University College London’s Institute of Clinical Trials and Methodology will provide its expert advice to EACTS as the new guidelines are developed.
- The following is the complete statement from SCAI:
"This week, leaders of the EACTS withdrew their support for European practice guidelines that endorse the use of coronary stents in many patients with left main coronary artery disease. EACTS leadership claims the results of the EXCEL randomized clinical trial were misleading because the protocol definition of myocardial infarction, based on the SCAI definition of myocardial infarction, did not align with the Third Universal Definition of Myocardial Infarction developed collaboratively by the European Society of Cardiology, the American College of Cardiology, the American Heart Association, and [World Heart Federation]. These definitions differ in the recommended biomarker measurements and thresholds for adjudicating a periprocedural myocardial infarction. EACTS leaders further expressed concern that mortality rates were greater among stent-treated patients in EXCEL, and suggested that the trial leadership was not forthcoming about this observation.
"The authors of the EXCEL trial preferred the SCAI definition of myocardial infarction because it is based on the best available evidence linking biomarker abnormalities to subsequent mortality in large clinical trials, avoids ascertainment bias, and uses the same criteria for PCI and bypass surgery. The SCAI definition has been used in multiple clinical trials because it avoids the pitfall of tabulating myocardial infarction events that are small enough to have little clinical impact. Instead, the SCAI definition permits assessment of myocardial events that are likely to be clinically relevant.
"Cumulative event curves for all-cause death for the two treatment groups in EXCEL were superimposable for the first 18 months after treatment, but separated from 18 to 36 months. Even though the difference was not statistically important at 3 years, it raised concerns that death rates would rise further for PCI-treated patients. For this reason, the recent publication of 5-year follow-up for EXCEL patients in The New England Journal of Medicine was welcomed. The paper shows that all-cause mortality remains higher for PCI-treated patients (13% vs 9.9%), but no difference in definite cardiovascular death was observed (5% vs 4.5%). The powered composite endpoint of death, stroke, or myocardial infarction was not different at 3 or 5 years of follow-up.
"SCAI respectfully disagrees with EACTS regarding EXCEL. While all-cause mortality is an important endpoint to follow, the EXCEL trial was neither designed nor powered to assess differences in this endpoint. Lethal differences in treatment effectiveness would be expected to manifest as differences in cardiovascular death rates; while this was also an unpowered observation, the lack of any difference in definite cardiovascular death at 5 years is reassuring. SCAI endorses the guidelines in Europe and the United States that support use of coronary stents for treatment of left main coronary artery disease in appropriately selected patients."