BOSTON — Adding fuel to the controversy of revascularization for left main disease, patients had better survival after coronary artery bypass grafting (CABG) surgery than percutaneous coronary intervention (PCI) in an all-comers nationwide registry study.
The incidence of all-cause mortality stayed greater up to 7 years after PCI versus CABG (adjusted HR 1.59, 95% CI 1.11-2.27) despite the use of various analytic methods to account for confounding in the Swedish Coronary Angiography and Angioplasty Registry (SCAAR), according to Elmir Omerovic, MD, PhD, of Sahlgrenska University Hospital in Gothenburg, Sweden.
Yet on subgroup analysis, there was significant interaction by age, such that people ages 80 and older lived longer after PCI for left main disease, whereas younger patients fared better with surgery, Omerovic reported at the Transcatheter Cardiovascular Therapeutics (TCT) meeting hosted by the Cardiovascular Research Foundation (CRF).
Results by restricted mean survival time suggested that patients who would be expected to live at least 7 years gained on average 6 months with CABG in lieu of PCI; for those with a shorter life expectancy, the estimated prolongation of life was less than 1 month.
Omerovic said the take-home message is that surgeons should not operate on high-risk older patients with left main disease, adding another point of discussion to the debate on stenting versus surgery for left main CAD.
While the 2016 NOBLE trial results favored CABG, the EXCEL trial concluded that there was no significant difference between strategies. Controversy erupted when a leading EXCEL investigator publicly broke away from the group in 2019 and cried foul over the trial’s unfair counting 0f periprocedural MIs that put surgery at a disadvantage.
Subsequently, in a meta-analysis pooling these two trials with SYNTAX and PRECOMBAT, investigators determined that there was no statistically significant difference in 5-year all-cause death between PCI and CABG, although a Bayesian analysis suggested slightly greater longevity after CABG.
“The question is not settled and I think we will continue to discuss it for years to come,” Omerovic said at a TCT press conference.
Omerovic reported that the 30 participating hospitals in SCAAR were roughly split between those performing more PCIs and those with more CABGs.
Included in the registry were all people in Sweden with stable angina, unstable angina, and non-ST-segment elevation MI (NSTEMI) who underwent coronary angiography since 2015 and had left main disease (>50% stenosis). Investigators found that these 10,254 individuals were roughly split between those who got PCI (52.6%) and those who underwent CABG (47.4%).
The two groups differed in various ways at baseline. For instance, people selected for PCI were a few years older and tended to present with more hyperlipidemia, previous MI, and prior revascularization procedures.
The prospect of selection bias was raised by Ori Ben-Yehuda, MD, CRF clinical trialist, who said he noticed an early separation of the curves that surprisingly favored CABG despite the immediate risk of surgical complications.
“Your curves open immediately and very wide … It’s a bit counterintuitive, ” agreed TCT session discussant Davide Capodanno, MD, of Azienda Ospedaliero-Universitaria Policlinico-Vittorio Emanuele in Catania, Italy.
Omerovic explained that the study relied on instrumental variable analysis — using each participating hospital’s preference for PCI or CABG as a source of randomness — and inverse probability weighting propensity scoring to help control for bias and confounding.
Despite these efforts, he acknowledged that selection bias and residual confounding cannot be excluded given the study’s observational nature.
“You have the instruments, the background, and the infrastructure to perform a randomized controlled trial exactly on this, and this is what needs to happen if you really want the answer,” commented TCT press conference panelist Roxana Mehran, MD, of Mount Sinai School of Medicine in New York City.
She added that cardiovascular mortality would be a helpful endpoint to assess in the setting of left main revascularization.
Omerovic and Ben-Yehuda disclosed no relationships with industry.
Capodanno disclosed relationships with Amgen, Arena, Daiichi-Sankyo/Eli Lilly, Sanofi-Aventis, and Terumo Medical, as well as an institutional relationship with Medtronic.
Mehran disclosed multiple relationships with industry.