TCT-100 Real-World Experience and Outcomes of Protected Versus Unprotected Left Main Percutaneous Coronary Intervention: Insights From the VA CART Program

Background: Outcomes of protected left main (PLM) and unprotected left main (ULM) percutaneous coronary intervention (PCI) are not well defined in contemporary U.S. practice. Previous studies of real-world data have shown worse in-hospital outcomes of ULM PCI compared with randomized trial data. We used a large national registry to characterize real-world practice and outcomes of left main PCI. Methods: Data were collected from the Veteran Affairs (VA) Clinical Assessment Reporting and Tracking (CART) Program for patients undergoing left main PCI between 2009 and 2019. PLM PCI was defined by the presence of at least 1 functioning bypass graft, and ULM PCI was defined as patients with no bypass grafting. Temporal trends, patient and procedure characteristics, anatomic complexity, and clinical complexity were assessed. A 1-to-1 propensity-matched analysis was performed using common comorbidities and clinical variables. One-year outcome analyses were conducted for major adverse cardiovascular events (MACE), all-cause mortality, rehospitalization for myocardial infarction (MI) and revascularization. Results: Of 4,351 patients undergoing left main PCI, 2,800 were PLM PCI and 1,551 were ULM PCI, of which 1,335 PLM and ULM PCI were included in the propensity matched cohort. Patients undergoing ULM PCI were older, more likely to present with acute coronary syndrome (ACS) and had a higher clinical complexity. In the propensity-matched cohort, there was no difference in age, rate of ACS presentation, burden of comorbidities, or left ventricular ejection fraction. There were no differences in in-hospital adverse events between the 2 groups. At 12 months, MACE occurred more frequently with ULM PCI compared with PLM PCI (25% [334] vs 20% [270]; P = 0.004), and all-cause mortality was also higher (18% [239] vs 14% [185]; P = 0.005). There was no difference in rehospitalization for MI, stroke, or revascularization at 12 months. Conclusion: In the VA Healthcare System, patients undergoing ULM PCI were older and more clinically complex than those undergoing PLM PCI. In the propensity-matched cohort, patients undergoing PLM PCI had better 12 outcomes than those undergoing ULM PCI, but there was a high rate of mortality and MACE at 1 year in both groups, despite a relatively low rate of MI or revascularization