Aims Observational data suggest that the use of bilateral internal mammary arteries (BIMA) during coronary artery bypass graft surgery provides superior revascularization to a single internal mammary artery (SIMA), but concerns about safety have prevented the widespread use of BIMA. The Arterial Revascularisation Trial (ART) is a randomized trial of BIMA vs. SIMA, with a primary outcome of survival at 10 years. This paper reports mortality, morbidity, and resource use data at 1 year. Methods and results Coronary artery bypass graft patients were enrolled in 28 hospitals in seven countries. Three thousand one hundred and two patients were randomly assigned to SIMA (n = 1554) or BIMA (n = 1548). The mean number of grafts was 3 for both groups. Forty per cent of the SIMA procedures and 42% of the BIMA were performed off-pump. Mortality at 30 days was 18 of 1548 (1.2%) for SIMA and 19 of 1537 (1.2%) for BIMA, and at 1 year was 36 of 1540 (2.3%) and 38 of 1529 (2.5%), respectively. The rates of stroke, myocardial infarction, and repeat revascularization were all =2% at 1 year and similar between the two groups. Sternal wound reconstruction was required in 0.6 and 1.9% of the SIMA and BIMA groups, respectively. Conclusion Data from ART demonstrate similar clinical outcomes for SIMA and BIMA at 1 year but BIMA grafts are associated with a small absolute increase (1.3%) in the need for sternal wound reconstruction. The results suggest that the use of BIMA grafts is feasible on a routine basis. The 10-year results of the ART will confirm whether BIMA grafting results in lower mortality and the need for repeat intervention.