Abstract OBJECTIVES We used individual patient data from 4 of the largest contemporary coronary bypass surgery trials to evaluate differences in long-term outcomes when radial artery (RA), right internal thoracic artery (RITA) or saphenous vein graft (SVG) are used to complement the left internal thoracic artery-to-left anterior descending graft. METHODS Primary outcome was all-cause mortality. Secondary outcome was a composite of major adverse cardiac and cerebrovascular events (all-cause mortality, myocardial infarction and stroke). Propensity score matching and Cox regression were used to reduce the effect of treatment selection bias and confounders. RESULTS A total of 10 256 patients (1510 RITA; 1385 RA; 7361 SVG) were included. The matched population consisted of 1776 propensity score-matched triplets. The mean follow-up was 7.9 ± 0.1, 7.8 ± 0.1 and 7.8 ± 0.1 years in the RITA, RA and SVG cohorts respectively. All-cause mortality was significantly lower in the RA versus the SVG [hazard ratio (HR) 0.62, 95% confidence interval (CI): 0.51–0.76, P = 0.003] and the RITA group (HR 0.59, 95% CI 0.48–0.71, P = 0.001). Major adverse cardiac and cerebrovascular event rate was also lower in the RA group versus the SVG (HR 0.78, 95% CI 0.67–0.90, P = 0.04) and the RITA group (HR 0.75, 95% CI 0.65–0.86, P = 0.02). Results were consistent in the Cox-adjusted analysis and solid to hidden confounders. CONCLUSIONS In this pooled analysis of 4 large coronary bypass surgery trials, the use of the RA was associated with better clinical outcomes when compared to SVG and RITA.