In Long-term, Real-world Data, CABG Bests PCI for Diabetic Patients With MVD

With practice changes over time, experts argue that a heart team approach is still best for choosing how to treat this cohort.

In Long-term, Real-world Data CABG Bests PCI for Diabetic Patients With MVD

Among patients with both diabetes and multivessel disease, long-term mortality and MACCE outcomes favor CABG surgery over PCI, according to new clinical and administrative registry data.

“In this particular setting of FREEDOM trial-type patients, I think the evidence is pretty strong that the first treatment of choice in these patients should be CABG surgery,” senior study author Douglas S. Lee, MD, PhD (Institute for Clinical Evaluative Sciences and Peter Munk Cardiac Centre, Toronto, Canada), told TCTMD. “There's biological reasons for that and clinical trial data supporting that and now a real-world effectiveness study also supporting that, so I think that's pretty strong evidence for CABG surgery in this type of patient.”

Indeed, CABG maintained a survival advantage over PCI for this cohort in the randomized FREEDOM trial at 8 years, but limitations of that analysis included substantial loss to follow-up as well as use of older drug-eluting stents.

The results have the same caveats, according to Sripal Bangalore, MD (NYU Langone Medical Center, New York, NY), who was not involved in the study. “I think we have more questions than answers,” he told TCTMD, pointing out that 23.3% of the patients here receive a bare-metal stent, which is “unheard of” today. “We really need a study which can address PCI versus CABG using latest PCI stents and techniques. So, until that time I think you really have to base it on a heart team approach, discussing risk versus benefit. I don't think the overwhelming benefit seen in this study is supported by randomized trials.”

The new analysis was published in the September 8, 2020, issue of the Journal of the American College of Cardiology and presented today as an oral abstract during the virtual European Society of Cardiology Congress 2020.

CABG Mortality, MACCE Advantage

For the study, Derrick Tam, MD, PhD (Sunnybrook Health Sciences Centre, Toronto, Canada), Lee, and colleagues retrospectively included all patients with diabetes and angiographic evidence of two- or three-vessel CAD who underwent either PCI (n = 4,519) or isolated CABG (n = 9,716) in Ontario, Canada, between 2008 and 2017.

A propensity score-matched analysis of 4,301 balanced pairs demonstrated no difference in 30-day mortality between the PCI and CABG cohorts (2.4% vs 2.3%; P = 0.721) or in-hospital stroke or MI. However, over a median follow-up period of 5.5 years, all-cause mortality (primary endpoint) was higher with PCI than CABG (HR 1.39; 95% CI 1.28-1.51), as was MACCE, the composite of MI, repeat revascularization, stroke, or death (HR 1.99; 95% CI 1.86-2.12).

There was no difference in late stroke between the revascularization options, but CABG also had an advantage with regard to late MI (HR 2.32; 95% CI 2.04-2.64) and late repeat revascularization (HR 3.65; 95% CI 3.24-4.34).

Findings from a secondary analysis of patients with left main disease as well as sensitivity analyses looking at patients with pre-PCI surgical consults, excluding those with ACS, including only those treated with BMS, and examining temporal trends of procedures were similar to the overall results.

“The results of our study were very similar to the FREEDOM trial results, despite the fact that these were unselected patients—it was not patients under the microscope of a clinical trial setting,” Lee commented, saying he was surprised by this similarity. “I guess it reflects the fact that the trial likely recruited a very good range of patients that reflected patients that we see in real-world practice.”

Not ‘One Size That Fits All’

In an accompanying editorial, Rishi Puri, MBBS, PhD (Cleveland Clinic, OH), and colleagues write that the findings provide “substantial supporting evidence that, with all other things being equal, CABG is the preferred revascularization strategy in patients with diabetes with multivessel CAD, with strong evidence now emanating both in the clinical trial and real-world settings.”

However, the totality of data supporting CABG over PCI may not be the final word given the evolution of various treatments over time, according to the editorialists. “Let us not forget the role of guideline-directed medical therapy (GDMT) in altering the disease milieu,” they write. “As background therapy, patients with diabetes and CAD should all receive high-intensity statins, angiotensin-converting enzyme inhibitors and/or angiotensin receptor blockers, and perhaps PCSK9 inhibitors. If one were to subject a population with diabetes and multivessel CAD to contemporary state-of-the-art aggressive GDMT, how would that potentially change the results of the analysis by Tam et al?”

Here, Lee said “the issue of revascularization using any approach versus guideline-directed medical therapy is a really good question especially in light of data such as that coming from the ISCHEMIA trial. From my clinical experience, I would say that when I revascularize patients with diabetes with multivessel disease, there's a high propensity to have cardiac outcomes such as MI and even cardiovascular death. And so my own personal bias is that because myocardium is not recoverable after infarction that we should be doing what it takes to try and preserve myocardium, but that's an open question and it's a very good one raised by the editorialists.”

Bangalore agreed. “Given all the recent advances in medical therapy for diabetes—we keep getting more and more data about how beneficial they are—I bet that the differences between PCI and CABG will even more become narrower,” he said, adding that the ongoing TUXEDO-2 study of contemporary PCI in diabetes that he’s co-chairing will likely provide more information soon.

Puri and colleagues point out that “both interventional cardiology and cardiac surgical techniques have evolved. Do we need a new RCT in patients with diabetes and multivessel CAD that incorporates a heart team approach, a background of optimal GDMT, and routine assessment of invasive coronary physiology (iFR [instantaneous wave-free ratio]/and fractional flow reserve), with subsequent intravascular ultrasound-guided PCI being used?” they ask.

While practice has changed, Lee said the effects of CABG over PCI observed so far are large enough that “there would need to be substantial improvements in technology to undo the findings from both the FREEDOM trial and from our large real-world evaluation. These major, practice-changing technologies could happen in the future, and so in the future, further evaluations/trials may be required. . . . As practice evolves though, it's always good to continuously evaluate how evolutions in practice are affecting outcomes, and so I think it's a healthy exercise to continue to evaluate these two revascularization strategies over time, especially if there are major technological advances.”

For now, the editorialists advise a patient-centered, shared decision-making approach with a heart team to make the call between CABG and PCI for this patient cohort. “Clearly, there is not ‘one size that fits all’ when treating such complex patients with diabetes and multivessel CAD,” they say.

  • This study was supported by a foundation grant from the Canadian Institutes of Health Research (CIHR).
  • Tam reports receiving support from a CIHR fellowship.
  • Lee reports receiving support by a Mid-Career Investigator Award from the Heart and Stroke Foundation and being the Ted Rogers Chair in Heart Function Outcomes, a joint University-Hospital Chair of University of Toronto and University Health Network.
  • Puri reports receiving consulting fees from Medtronic, Amgen, and Cerenis as well as holding minor equity in Centerline Biomedical.