Outcomes Following Percutaneous Coronary Intervention in Patients With Multivessel Disease Who Were Recommended for But Declined Coronary Artery Bypass Graft Surgery

Background Patients may prefer percutaneous coronary intervention (PCI) over coronary artery bypass graft (CABG) surgery, despite heart team recommendations. The outcomes in such patients have not been examined. We sought to examine the results of PCI in patients who were recommended for but declined CABG. Methods and Results Consecutive patients with stable ischemic heart disease and unprotected left main or 3‐vessel disease or Synergy Between Percutaneous Coronary Intervention With Taxus and Cardiac Surgery score >22 who underwent PCI after heart team review between 2013 and 2020 were included. Patients were categorized into 3 groups according to heart team recommendations on the basis of appropriate use criteria: (1) PCI‐recommended; (2) CABG‐eligible but refused CABG (CABG‐refusal); and (3) CABG‐ineligible. The primary end point was the composite of death, myocardial infarction, or stroke at 1 year. The study included 3687 patients undergoing PCI (PCI‐recommended, n=1718 [46.6%]), CABG‐refusal (n=1595 [43.3%]), and CABG‐ineligible (n=374 [10.1%]). Clinical and procedural risk increased across the 3 groups, with the highest comorbidity burden in CABG‐ineligible patients. Composite events within 1 year after PCI occurred in 55 (4.1%), 91 (7.0%), and 41 (14.8%) of patients in the PCI‐recommended, CABG‐refusal, and CABG‐ineligible groups, respectively. After multivariable adjustment, the risk of the primary composite outcome was significantly higher in the CABG‐refusal (hazard ratio [HR], 1.67 [95% CI, 1.08–3.56]; P=0.02) and CABG‐ineligible patients (HR, 3.26 [95% CI, 1.28–3.65]; P=0.004) groups compared with the reference PCI‐recommended group, driven by increased death and stroke. Conclusions Cardiovascular event rates after PCI were significantly higher in patients with multivessel disease who declined or were ineligible for CABG. Our findings provide real‐world data to inform shared decision‐making discussions.