乌鸦传媒

Operators couldn't achieve surgery-like results with percutaneous coronary intervention (PCI) for three-vessel coronary artery disease (CAD) even with the help of newer stents and physiological guidance, according to results of the FAME 3 trial.

A strategy of contemporary PCI guided by fractional flow reserve (FFR) did not meet noninferiority against coronary artery bypass grafting (CABG), as 10.6% versus 6.9% of patients, respectively, had died or suffered a myocardial infarction (MI), stroke, or repeat revascularization within a year (HR 1.5, 95% CI 1.1-2.2), according to William Fearon, MD, of Stanford University School of Medicine in California.

Removing repeat revascularization from the primary endpoint still gave CABG the advantage, albeit it was no longer statistically significant (7.3% with PCI vs 5.2% with CABG; HR 1.4, 95% CI 0.9-2.1), Fearon told the audience during a late-breaking trial session at the (TCT) meeting. The FAME 3 results were also published simultaneously online in the .

With this trial, CABG maintains its longstanding advantage over PCI in multivessel disease (not counting left main disease).

In the time since the SYNTAX trial was conducted in 2005-2007, however, "what's remarkable is both arms did significantly better in terms of their outcomes," Fearon said during a TCT press conference. He suggested that part of this may be explained by improved adherence to medical therapy in the current era and improvements in surgical operating technique.

"The low use of intravascular imaging to optimize PCI results is a real opportunity for PCI to get better. Unfortunately its use remains low, and that is one likely reason (among others too) why in contemporary practice CABG is better than PCI for [three-vessel disease]," wrote Ajay Kirtane, MD, of NewYork-Presbyterian/Columbia University Irving Medical Center in New York City, in the virtual chat discussion. Kirtane was not involved with the study.

Notably, the margin of noninferiority for PCI in FAME 3 had been generous from the start, and was widened even further during the trial's conduct: investigators had started with a 1.45 upper boundary in the 95% CI for the HR and raised it to 1.65 on the basis of low event accrual.

"I think the message is we are getting very good results with PCI, but better results with surgery ... It will be important to look at long-term follow-up and the impact of completeness of revascularization," commented Ran Kornowski, MD, of Rabin Medical Center in Petah Tikva, Israel, during the panel discussion.

"There are still patient tradeoffs. Some patients will accept the higher event rate in order to avoid a longer recovery and vice versa. [The study] allows patients to make more informed decisions," Fearon maintained.

was conducted at 48 centers in Europe, North America, Australia, and Asia. Eligible patients were candidates for either PCI or CABG, according to the local heart team. Investigators randomized 1,500 people with three-vessel CAD (excluding left main involvement) to one of the two procedures.

The overall FAME 3 cohort averaged just over 65 years of age. Over 80% were men, and over 90% were white. Nearly 40% presented with acute coronary syndrome.

The PCI group underwent angioplasty with a Medtronic Resolute drug-eluting stent if FFR fell below 0.80. FFR was measured in 82% of lesions, of which 24% turned up a positive result. PCI recipients got a mean 3.7 zotarolimus-eluting stents per person.

CABG operators could use FFR data for guidance if they so desired, and they did so in approximately 10% of cases. Patients in the surgical group underwent 3.4 distal anastomoses.

Patients with low SYNTAX scores (<23) represented a third of the group, those with intermediate scores (23-32) half, and those with high scores (>32) less than one-fifth.

Fearon reported that on subgroup analysis, people with less complex disease according to SYNTAX tended to fare better in avoiding major adverse cardiac or cerebrovascular events after PCI in relation to CABG. Yet this finding can only be hypothesis-generating because the trial was not powered for this analysis, he cautioned.

The study authors also acknowledged that they relied on a fairly restrictive definition of procedural MI. A more liberal definition would have resulted in more MIs for the CABG group, the researchers suggested from a post hoc analysis.

Additionally, despite the FFR guidance in the PCI arm, data on the completeness of revascularization are not available for either study arm.

"We were hoping to achieve functionally complete revascularization. There may have been some lesions that were not significant that were deferred. We found that was more common with the low SYNTAX group," Fearon noted.

He said that the FAME 3 team is working on a quality-of-life analysis for presentation next spring.

Comment