WHAT KINDS OF PATIENTS WITH CORONARY ARTERY DISEASE ARE TREATED WITH MULTIVESSEL PERCUTANEOUS INTERVENTION VERSUS BYPASS SURGERY? A PROVINCE-WIDE FIELD EVALUATION IN QUÉBEC HOSPITALS WITH ON-SITE CARDIAC SURGERY

Abstract

BACKGROUND: Wide variation across hospitals in choice of revascularization treatment has been observed and outcomes of percutaneous coronary intervention (PCI) versus coronary artery bypass surgery (CABG) are increasingly being examined. Our cardiology evaluation unit was mandated by the Quebec Ministry of Health to evaluate the practice of multivessel revascularization and its outcomes across Quebec’s 8 tertiary cardiac hospitals offering both PCI and CABG. METHODS: We identified all patients who underwent either multivessel ( 2 myocardial territories) PCI or CABG in 2010-12. A maximum of 300 patients treated with CABG and 300 patients treated with PCI (excluding primary PCI) per hospital were randomly selected for chart review. Factors associated with choice of treatment (PCI vs CABG) were identified using multivariate regression analysis. RESULTS: The study cohort included 2016 PCI patients and 2274 CABG patients. At the provincial level, factors associated with treatment by PCI included older age ( 75 years; OR1⁄41.7; 95% CI: 1.4-2.1), female sex (OR1⁄41.8; 95% CI: 1.4-2.2) and previous PCI (OR1⁄42.1; 95% CI: 1.7-2.5). Relative to patients with elective status, those with urgent/ emergent status were more likely to have undergone PCI than CABG (OR1⁄42.0; 95% CI: 1.6-2.4). Factors associated with CABG were arterial hypertension (OR1⁄41.7; 95% CI: 1.42.1), heart failure (OR1⁄43.3; 95% CI: 2.5-4.3) and 3-vessel disease (OR1⁄416.7; 95% CI: 13.9-20.0). Across the 8 hospitals, the proportion of female patients aged 75 ranged from 31 to 54% for both PCI and CABG. The proportion of elective procedures across hospitals ranged from 17% to 35% for PCI and from 26% to 48% for CABG. The proportion of patients treated with PCI for 3-vessel disease varied by hospital from 5% to 19% while it ranged from 60% to 78% for CABG. Diabetes was not associated with choice of treatment. Across the 8 hospitals, prevalence of diabetes varied from 23% to 36% among PCI patients and from 31% to 40% among CABG patients. One in 9 multivessel PCI patients underwent unprotected left main intervention (range 2.5% to 15.2% across hospitals). CONCLUSION: Although CABG was the predominant treatment for 3-vessel disease, use of PCI in these patients was not uncommon and varied widely by hospital. Presence of diabetes did not impact treatment choice. PCI was used more frequently than CABG for non-elective cases. Patient characteristics varied more by hospital for PCI than for CABG. These results point to the importance of appropriate risk adjustment when comparing outcomes by choice of treatment and by hospital.