Conduit Choice and Volume-Outcome Relationships in Multiarterial Coronary Artery Bypass Grafting of Medicare Beneficiaries in the United States.

Overview

abstract

BACKGROUND: Multiarterial grafting is underused. We evaluated the association between surgeon experience and conduit use outcomes after multiarterial grafting. METHODS: Using US Centers for Medicare & Medicaid data, we identified 29,268 patients ≥65 years undergoing isolated, primary, nonemergency multiarterial grafting, of whom 15,682 met the inclusion criteria. Propensity score matching was performed on 26 baseline patient characteristics. Individual surgeons were grouped into tertiles on the basis of conduit use: low volume, <3 radial arteries or 2 bilateral internal thoracic artery (BITA) grafts annually; high volume, >10 radial arteries or 4 BITA grafts annually. The primary outcome was major adverse cardiac events (MACE): death, myocardial infarction, or repeated revascularization at 4 years. Outcomes were compared in a multivariable Cox proportional hazards model adjusting for the individual surgeon's case volume of each conduit. RESULTS: Among recipients of multiarterial grafting, 5784 (20%) received radial artery grafts and 9898 (34%) received BITA grafts. Radial artery recipients were younger (70.6 vs 71.2 years) and more likely to be diabetic (46.9% vs 43.2%) than BITA recipients. Of 5778 matched pairs, at 4 years, the incidence of MACE was numerically lower in the radial artery group (14.7% vs 15.7%; P = .05), but there was no difference in all-cause mortality (10.8% radial artery vs 11.5% BITA; P = .06). For BITA graft recipients, surgeon experience was associated with MACE only in the lowest vs the highest volume tertile (adjusted hazard ratio, 1.15; 95% CI, 1.01-1.33; P = .046). There was no association between MACE and surgeon volume for radial artery graft recipients. CONCLUSIONS: Radial artery and BITA grafting demonstrated similar midterm outcomes, whereas there was a surgeon-volume effect for BITA use.