Left-Main PCI, CABG Mortality Similar; Choice Remains Complex

November 17, 2021

Cardiology has its share of controversies. Although some of them are settled, others just settle down while the field does its current best for patients and ponders how the next clinical trial should approach unresolved questions.

The latter outcome appears likely for at least one protracted, sometimes fiery debate. At issue: For most patients with low-to-intermediate complexity left-main coronary disease, is the preferred procedure coronary stenting or bypass surgery?

"Yes" is the best answer for now, suggests a new meta-analysis led by a group of highly respected researchers and published in a premier journal.

There were no significant differences in mortality or stroke risk at 5 years between percutaneous coronary intervention (PCI) with drug-eluting stents or coronary bypass surgery (CABG) for left-main disease in the patient-level analysis based on four major randomized trials going back more than a decade.

But broader secondary results reflect the issue's complexity. Strokes at 1 year were fewer with PCI, as were periprocedural myocardial infarctions (MI). On the other hand, a Bayesian analysis gave a likely but slight 5-year survival edge to CABG, which also did better overall for MI and repeat revascularization.

In that Bayesian analysis, the probability that 5-year mortality was higher with PCI than with CABG was as high as 86%. There was a 49% probability that the absolute excess PCI mortality was at least 1%, and a 5.2% probability that it was higher than 2.5%.

"In other words, the difference was, more likely than not, less than 0.2% per year," said Marc S. Sabatine, MD, MPH, when presenting the meta-analysis at the virtual American Heart Association (AHA) Scientific Sessions 2021.

Sabatine, of Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, chairs the venerable Thrombolysis in Myocardial Infarction (TIMI) Study Group and is lead author on the meta-analysis published November 15 in The Lancet to coincide with his presentation.

The results apply to patients like those in the four trials making up the meta-analysis — SYNTAX, PRECOMBAT, NOBLE, and EXCEL — who had mostly low-to-intermediate-risk coronary anatomy and had been deemed suitable for either PCI or CABG.

"For any given patient, our findings need to be put into the context of factors that must be taken into account when deciding on a revascularization strategy," Sabatine said. "The complex nuances of these data emphasize the importance of a heart-team approach to assist patients in reaching a treatment decision that is best for them."

An accompanying editorial had a different take on the Bayesian survival analysis. It was a limitation of the study "that it focused only on superiority, which was not established for PCI," contend Emilie P. Belley-Côté, MD, PhD, and Philip J. Devereaux, MD, PhD, McMaster University, Hamilton, Ontario, Canada.

The meta-analysis authors, they write, "did not attempt to directly inform the noninferiority of PCI relative to CABG by setting a noninferiority boundary; however, physicians and patients can draw their own conclusions given the 49.1% probability that PCI compared with CABG had a 1% or higher absolute excess risk of death at 5 years."

Moving the Debate Forward

The meta-analysis was "rigorously" conducted and "will be extremely useful to heart teams and to patients engaged in shared-decision making," Deepak L. Bhatt, MD, MPH, Brigham and Women's Hospital Heart & Vascular Center, Boston, told theheart.org | Medscape Cardiology.

"Detailing the different trade-offs between PCI and CABG, as this paper nicely does, really moves forward the debate on this contentious issue in an evidence-based, constructive way that will ultimately benefit patients," said Bhatt, who is not a coauthor on the report.

The study "seems to have been done in a very upfront, transparent fashion," and was led by clinicians who are neither surgeons nor interventional cardiologists and so would be seen as professionally impartial, said Sunil V. Rao, MD, Duke Clinical Research Institute and Duke University Hospital, Durham, North Carolina.

Speaking with theheart.org | Medscape Cardiology, Rao referred to recent heated public discussions — especially on social media — regarding alleged improprieties and other issues about how the EXCEL trial was conducted and analyzed that seemed to turn the issue into a contest between the two revascularization approaches. "Which one is the winner, PCI or CABG? I don't think that's really the way clinical medicine should work," he said.

"I'm hoping that cooler heads will prevail after this, and we realize that it's not about whether the surgeons win or the interventionalists win," Rao said. "It's really, how do we best help the patient win?" The meta-analysis "shows us that the only way to get there is really through collaboration."

It's critical not to frame the issue as a competition between PCI and CABG to establish exclusive use in left-main coronary disease, agreed Robert Harrington, MD, Stanford University, California, as a discussant for Sabatine's presentation.

The meta-analysis points to what has been apparent about PCI and CABG "since the very beginning." The optimal choice in left-main disease "depends upon the patient's values and preferences as to what they're trying to avoid and what they're trying to achieve," Harrington said. "It really does require careful conversation, careful consideration of an individual patient's clinical and anatomic characteristics."

The findings didn't tilt the evidence clearly toward either procedure, and "I think we still have opportunities to further explore this question as to a preferred therapy," he said. "But these data provide us really great quantitative information for what is the essence of patient-based decision-making."

"I think the data in this paper really do inform us," Rao agreed. "I think you can use them to have a conversation with the heart team and the patient, to figure out how a particular patient with left-main disease, with or without other coronary disease, would prefer to be treated."

Procedural vs Spontaneous MI

The meta-analysis combined 4394 patients with left-main disease from the four trials, with 1905 — the most from any one study — hailing from EXCEL.

Estimated all-cause mortality at 5 years, its primary endpoint, was 11.2% for PCI and 10.2% for CABG, for a hazard ratio (HR) of 1.10 (95% CI, 0.91 - 1.32; P = .33).

Endpoints favoring CABG included spontaneous MI (HR, 2.35 [95% CI, 1.71-3.23; P < .0001]), with the rates differing 3.5 absolute percentage points; and repeat revascularization (HR, 1.78 [95% CI, 1.51 - 2.10; P < .0001]), with an absolute difference of 7.6%.

How PCI and CABG compared for procedural MI depended on whether the endpoint was defined per each study's protocol or according to the third or fourth Universal Definition of Myocardial Infarction (UDMI).

Under the protocol definitions, PCI entailed a lower procedural MI risk (odds ratio (OR), 0.65 [95% CI, 0.47 - 0.92; P = .013]). The risk flipped using UDMI criteria, but without reaching significance, with an OR of 1.42 (0.88 - 2.30; P = .15).

From baseline to 1 year, Sabatine said, "there was a lower rate of stroke with PCI, with almost a two-thirds reduction in risk, and an absolute risk difference of 1%." The 1-year HR was 0.37 (0.19 - 0.69). "Convincingly, this pattern was seen in all four trials."

But stroke risks leveled out over 5 years, with an HR for PCI of 0.84 (95% CI, 0.59 - 1.21).

The data upon which the patient-level meta-analysis is based suggest that "the totality of patients randomly assigned across all the trials to date is insufficient to clearly establish superiority or noninferiority of PCI over the current gold standard of CABG," editorialists Belley-Côté and Devereaux write. "A trial larger than the aggregate of SYNTAX, PRECOMBAT, NOBLE, and EXCEL should be a priority in the field."

The meta-analysis received no external funding. Sabatine and other authors are members of the TIMI Study Group, which has received institutional grant support from Abbott, Amgen, Anthos Therapeutics, AstraZeneca, Bayer HealthCare Pharmaceuticals, Daiichi-Sankyo, Eisai, Intarcia, Ionis, MedImmune, Merck, Novartis, Pfizer, Quark Pharmaceuticals, Regeneron Pharmaceuticals, Roche, Siemens Healthcare Diagnostics, The Medicines Company, and Zora Biosciences. Sabatine himself disclosed grant support through his institution from Abbott, Amgen, Anthos Therapeutics, AstraZeneca, Bayer, Daiichi-Sankyo, Eisai, Intarcia, Ionis, The Medicines Company, MedImmune, Merck, Novartis, Pfizer, and Quark Pharmaceuticals; and consulting for Althera, Amgen, Anthos Therapeutics, AstraZeneca, Bristol Myers Squibb, CVS Caremark, DalCor, Dr Reddy's Laboratories, Fibrogen, IFM Therapeutics, Intarcia, MedImmune, Merck, Moderna, Novo Nordisk, and Silence Therapeutics. Bhatt and Rao are interventional cardiologists. Rao has disclosed no relevant financial relationships. Bhatt has disclosed grants and/or personal fees from many companies; personal fees from WebMD and other publications or organizations; and having other relationships with Medscape Cardiology and other publications or organizations. Belley-Côté disclosed grant funding from Bayer, Bristol Myers Squibb–Pfizer, and Roche Diagnostics. Devereaux reported grants to his institution from Abbott Diagnostics, Roche Diagnostics, and Siemens; receiving a product for a clinical trial at his institution from Philips Healthcare; and participating in an advisory board meeting for Bayer and Quidel.

American Heart Association (AHA) Scientific Sessions 2021: Session FS.06. Presented November 15, 2021.

Lancet. Published online November 15, 2021. Article, Editorial

Follow Steve Stiles on Twitter: @SteveStiles2. For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.