Registries Point to Lower Risk of Death With CABG vs PCI in Multivessel CAD

Patients with multivessel coronary artery disease undergoing CABG surgery are more likely to be alive at 1 year, and less likely to be readmitted to the hospital or require repeat revascularization when compared with individuals undergoing PCI, according to the results of a new analysis.

One-year mortality among the surgical patients was 7.2% compared with 11.5% among the PCI patients (P < 0.001), and this survival benefit with CABG surgery was observed in virtually all subgroups, say investigators.

“SYNTAX and FREEDOM, and other randomized trials that have looked at the treatment of patients with multivessel disease, have been incredibly important,” said lead investigator Suresh Mulukutla, MD (University of Pittsburgh Medical Center, PA). “In the drug-eluting stent era, they completely changed how we approach patients who have multivessel disease, but in an era when a lot of patients get treated with PCI now, are the findings we see in studies like SYNTAX and FREEDOM replicable in a real-world, routine-care environment?”

In Europe, the clinical guidelines favor CABG surgery over PCI, with CABG a class I indication in patients with complex or moderately complex multivessel CAD. PCI is a class III indication in patients with complex anatomy, but both CABG and PCI are class I indications in patients with multivessel CAD of low complexity. In the United States, surgery is a class I indication for three-vessel CAD while PCI is a class IIb indication. CABG is preferred over PCI in patients with moderately or highly complex anatomy.

To TCTMD, Mulukutla said that both CABG surgery and PCI are reasonable options in multivessel CAD, depending on the patient’s anatomy and characteristics, but because of the less invasive nature of PCI, that strategy is understandably preferred by patients. Their analysis, he said, was to determine if this “preference” was associated with adverse outcomes in the multivessel disease population treated in the contemporary era.

Which Way to Revascularize Multivessel CAD Patients?

In the SYNTAX multivessel cohort, which compared CABG versus PCI in patients with three-vessel CAD, the 5-year results showed that surgery was associated with a lower risk of death, MI, and coronary revascularization compared with PCI, particularly among intermediate- and high-risk patients. The FREEDOM trial differed in that investigators included patients with diabetes and multivessel disease, and long-term results showed a definite advantage with surgery over percutaneous revascularization. An older observational analysis of patients treated in New York State also showed a survival advantage with CABG surgery over PCI in patients with multivessel disease.

However, several observational analyses suggested comparable short- and long-term mortality benefits with CABG and PCI using latest-generation everolimus-eluting stents (Xience; Abbott Vascular). Additionally, a recent patient-level, pooled analysis of 11 randomized trials comparing CABG with PCI in patients with multivessel or left main disease showed that all-cause mortality at 5 years was significantly lower with CABG surgery in patients with diabetes, but not in those without the metabolic condition.   

“[Ours] is clearly a retrospective, observational analysis, and it’s very important when we look at these questions that we don’t go into it with preconceived notions of what the result should be,” said Mulukutla. “So, when we see discrepant data about a mortality difference [that] we’re seeing in this real-world analysis but don’t see the mortality benefit in the randomized trials, I actually think that’s incredibly important to recognize and to understand why the results are different.”

The present study, published online today in the Annals of Thoracic Surgery, included an initial 6,163 patients with multivessel CAD in the National Cardiovascular Data Registry (NCDR) or Society of Thoracic Surgeons (STS) database who underwent coronary revascularization between 2010 and 2018. All patients in the surgical arm underwent isolated CABG, while those in the PCI arm were included if they had three-vessel CAD defined by 70% or greater stenosis in all three major coronary arteries, left main coronary stenosis 50% or greater, or two-vessel CAD defined by stenoses 70% or greater in two major vessels, including the LAD. For those who underwent PCI, everolimus-eluting and zotarolimus-eluting stents were used in 66% and 30% of patients, respectively. 

In the propensity-matched analysis, which included 844 patients in each group, PCI was associated with a significantly higher risk of mortality at 1 year (HR 1.64; 95% CI 1.29-2.10). One-year readmission rates were 38.4% and 28.1% in the PCI and CABG groups, respectively (P < 0.001), while freedom from repeat revascularization was also significantly lower among the surgical patients (6.7% with PCI vs 1.0% with surgery; P < 0.001).

Makes Sense Physiologically

The investigators stress that patients in the NCDR and STS databases were older, had more comorbid conditions, and were more like to have NSTEMI (27.8%), which may explain why their results differ from some of the randomized controlled trials showing no survival difference with CABG. 

“We were surprised how stable the benefit was across the different subpopulations,” Mulukutla said about their findings. “The mortality benefit occurred early at 1 year and it was sustained in diabetic patients, nondiabetic patients, in patients with normal LV function, and in patients with depressed LV function. That was also a little bit of surprise to us, just how marked and sustained that benefit was [in the subgroups].”

Todd Rosengart, MD (Baylor College of Medicine, Houston, TX), a cardiothoracic surgeon who was not involved in the study, said these latest results are consistent with other real-world data, such as results from the New York State database.

“To us, physiologically it makes sense, as we know clearly that CABG tends to do better than PCI in terms of complete revascularization, and complete revascularization leads to improved survival,” he told TCTMD. “I think there’s been some hand-waving around the concept of treating the lesion versus perfusing the distal myocardium, and CABG is able to provide blood flow to the entire myocardium as opposed to just attacking a single lesion, which is prone to reocclusion. So this very much fits with our real-world experience.”

Interventional cardiologist Paul Fiorilli, MD (Penn Medicine, Philadelphia), who was also not involved the study, agreed that these latest findings align with data from the other real-world, observational studies. One of the limitations of observational analyses, however, is the absence of detailed patient information that would allow comparisons between CABG and PCI in patients with varying CAD complexity.

“One of the factors here that is interesting is that we don’t see a breakdown by severity of coronary artery disease and SYNTAX score, factors that we know are associated with triaging patients to better outcomes with one revascularization over another,” said Fiorilli.  

Heart Team Approach for Multivessel CAD

Like the investigators, Fiorilli noted there has been a shift towards greater use of PCI, even in multivessel CAD, and said this shift is largely driven from patient preference and their concerns about open-heart surgery.

“I think this study really reinforces the need for a comprehensive heart team approach in these multivessel coronary artery disease patients so we can fully weigh the benefits and risks of each strategy, keeping in mind the patient’s perspective,” Fiorilli told TCTMD. “Often these decisions get made on the fly in the cath lab when you see multivessel disease. It can sometimes be tough to get a surgical opinion in a quick fashion, but I think it stresses the importance of stopping, and asking ourselves what is the optimal strategy for revascularization for this particular patient.”    

Mulukutla told TCTMD their results should not be construed to mean all patients with multivessel CAD should be undergoing CABG surgery, but rather that providers need to consider both revascularization strategies as viable options. “I would argue for the heart team approach for the care of patients with multivessel disease so that there is more than one voice being heard, and to take into account the various aspects of the patient’s clinical situation,” he said. “We should allow patients the opportunity to have that informed consent.”

Rosengart also agreed with the need for a heart team approach in the setting of multivessel CAD, noting that most patients will shy away from a procedure where “they crack the chest open” as opposed to the percutaneous approach. “Patients naturally think of PCI as being better because it’s safer, but these data, and all of the other data in this regard, suggest the opposite,” he said. “PCI is certainly less taxing in the short run on the patient—shorter hospital stays, quicker recovery, et cetera—but we as surgeons look at the short-term investment of a little more invasive procedure to realize significant benefits.”

The heart team with a cardiologist and cardiac surgeons discussing options with the patient can help individuals make fully informed decision. “Even if they don’t say the same thing, at least the patient can hear both parties in the same room, at the same time, sharing their different perspectives,” said Rosengart.