Long-Term Survival After On-Pump and Off-Pump Coronary Artery Bypass Grafting

Abstract

Background: Off-pump coronary artery bypass grafting (CABG) may be associated with increased hazard for long-term mortality as compared with on-pump CABG. We sought to evaluate risk-adjusted long-term survival after off-pump and on-pump CABG, particularly among high-volume and low-volume CABG surgeons.

Methods: We evaluated 1,235,089 isolated CABGs (off pump = 209,085; on pump = 1,026,004) performed in Medicare beneficiaries from 2001 to 2015. Long-term hazard for mortality after off-pump versus on-pump CABG was compared with Kaplan-Meier and log-rank analysis among all CABG surgeons as well as among high-volume and low-volume CABG surgeons, before and after inverse probability of treatment weighting to adjust for confounding.

Results: Among all surgeons, off-pump CABG was associated with a statistically significant hazard for mortality as compared with on-pump CABG before and after inverse probability of treatment weighting (median survival: off pump 9.8 years vs on pump 10.2 years; difference in median survival -134 days; log-rank P < .001). Cox regression analysis confirmed an interaction between surgeon volume and long-term mortality. The hazard for mortality associated with off-pump CABG was decreased among high-volume surgeons (difference in median survival -84 days; log-rank P < .001) and increased among low-volume surgeons (difference in median survival -240 days; long-rank P < .001).

Conclusions: Off-pump CABG was associated with a significant, but clinically modest, increased hazard for mortality as compared with on-pump CABG. The hazard was reduced when off-pump CABG was performed by high-volume CABG surgeons.