METHODS: We analysed prospectively collected data from the multicentre IRIS-MAIN registry, conducted at 46 hospitals in South Korea, India, Japan, Indonesia, and Malaysia. The registry enrolled adults aged 18 years or older with unprotected left main coronary artery disease (defined as a stenosis diameter of ≥50% as visually estimated on angiography), who had been treated with PCI, CABG, or medical therapy alone. The primary outcome was all-cause mortality. We used a target trial emulation framework to compare the long-term risk of all-cause mortality between PCI with drug-eluting stents and isolated CABG. We used propensity-score overlap weighting to adjust for measured baseline covariates. The IRIS-MAIN Registry study is registered with ClinicalTrials.gov (NCT01341327).

FINDINGS: Between Jan 1, 2003, and Dec 31, 2024, 6900 patients were enrolled in the IRIS-MAIN registry, of whom 5379 (78·0%) underwent coronary revascularisation through either PCI with drug-eluting stents (3836 [71·3%]) or CABG (1543 [28·7%]) and were included in the analysis. The median duration of follow-up for all patients was 10·0 years (IQR 5·0-16·7). In the overlap-weighted population, no statistically detectable difference in 10-year weighted Kaplan-Meier rates of all-cause mortality was observed between PCI and CABG (24·0% vs 22·9%; overlap-weighted hazard ratio 1·00 [95% CI 0·87-1·14]; p=0·99). The confidence interval does not exclude a mortality benefit or harm with either strategy.

INTERPRETATION: In this non-randomised registry analysis of patients with left main coronary artery disease, there was no statistically detectable difference in all-cause mortality between PCI and CABG. For patients considered suitable for either strategy, multidisciplinary Heart Team discussion might help patients and clinicians weigh potential trade-offs in the choice between PCI and CABG. These findings are observational and cannot establish causal treatment effects or equivalence in survival.