Abstract / Summary
Background: Optimal stenting for distal unprotected left main coronary artery (LMCA) bifurcation disease remains uncertain. We evaluated long-term outcomes by initially intended strategy. Methods: This retrospective, single-centre cohort included 234 consecutive patients undergoing percutaneous coronary intervention (PCI) without acute coronary syndrome, classified by intended strategy irrespective of crossover. Cox models for all-cause mortality were adjusted for strategy, age, left ventricular ejection fraction, SYNTAX score, and diabetes, with anatomical and propensity-score sensitivity analyses. Secondary and subgroup analyses were exploratory. Results: Provisional and upfront two-stent strategies were intended in 145 and 89 patients. Two-stent patients had greater anatomical complexity. During median 49-month follow-up, mortality did not differ significantly (two-stent versus provisional: 16.9% versus 21.4%; p = 0.499; adjusted hazard ratio, 0.83; 95% confidence interval, 0.43–1.60; p = 0.579), with consistent sensitivity analyses. Older age, lower ejection fraction, and higher SYNTAX score were associated with mortality. Target-lesion revascularisation (7.8% versus 2.1%; p = 0.033) and in-hospital stroke (5.6% versus 0.7%; p = 0.031) were more frequent with intended two-stent PCI; other secondary outcomes did not differ significantly. Exploratory unadjusted comparisons showed lower target-lesion and target-vessel revascularisation with double-kissing crush (n = 74) versus conventional culotte (n = 15), limited by small samples and selection bias. Intravascular ultrasound was associated with lower mortality in exploratory adjusted analyses. Mortality was numerically lower with elective intra-aortic balloon pump support at ejection fraction ≤40%, without statistical significance. Conclusions: No, statistically significant association between initially intended bifurcation strategy and long-term all-cause mortality was observed. Limited precision precludes an inference of equivalence, and exploratory differences in target-lesion revascularisation and in-hospital stroke warrant caution. These findings provide a long-term real-world perspective on individualised distal LMCA PCI, integrating lesion complexity, ventricular reserve, procedural technique, and intravascular imaging.