Controversy exists regarding ideal approaches in teaching residents complex and/or new surgical techniques in part because consequences on patient outcomes are largely unknown. This study compared patient outcomes for cases in which residents (rather than attending surgeons) performed most of the distal anastomoses as primary surgeons, during on- and off-pump coronary artery bypass grafting (CABG).This preapproved substudy of the Randomized On/Off Bypass (ROOBY) trial compared clinical outcomes and 1-year graft patency for cases in which residents versus attending surgeons were the primary operator. Comparisons were made between on-pump and off-pump techniques.From July 2003 through May 2007, a total of 1272 ROOBY nonemergent CABG patients were randomized at 16 Veterans Affairs centers where residents were active participants. Residents were the primary surgeon (ie, performed ≥50% of the distal anastomoses) more frequently in on-pump (77.9%) than in off-pump (67.4%) cases. Between these 2 techniques, no differences were found [corrected] in baseline patient characteristics; short-term and 1-year morbidity and mortality rates were no different for residents versus attendings in CABG cases. FitzGibbon A graft patency rates were similar for resident versus attendings completed distal anastomoses for on-pump (83.0% vs 82.4%) compared with off-pump (77.2% vs 76.6%) procedures.In the ROOBY trial, short-term and 1-year patient outcomes and graft patency rates did not differ between resident and attending surgeons, demonstrating that with appropriate patient selection and resident supervision, residents can perform advanced, novel surgical techniques with outcomes similar to those of attending surgeons.