It seemed like a great idea. Perform coronary-artery bypass grafting (CABG) without stopping the heart and you can prevent the dangers of cardiopulmonary bypass. But it didn't work out. After growing in popularity over the previous decade, in 2009 the first large, well-conducted trial comparing on- and off-pump CABG -- the VA-sponsored Randomized On/Off Bypass (ROOBY) trial, with 2,203 patients enrolled -- pretty much dashed hopes for the less invasive procedure. At 30 days there were no significant differences between the two groups in the rate of death or complications. At 1 year, the combined rate of death from any cause, repeat revascularization procedure, or nonfatal MI was significantly higher in the off-pump than the on-pump group. "The most likely explanation for the ROOBY trial's unfavorable findings regarding off-pump CABG appears to center on comparative differences in complete and durable revascularization," wrote Eric Peterson (Duke University) in an accompanying editorial. Now the ROOBY investigators report the 5-year follow-up study outcomes in a paper published in the New England Journal of Medicine with further evidence that off-pump CABG is inferior to on-pump CABG. Although the difference in mortality by itself had not been significant at 1 year, it was at 5 years. Off-pump CABG resulted in a significant 28% increase in the risk of death at 5 years. The mortality rate was 15.2% in the off-pump group versus 11.9% in the on-pump group (P=0.02). The rate of major adverse cardiovascular events was 31% versus 27.1% (P=0.046). There were no advantages associated with off-pump CABG. "The 5-year outcomes in our study support the conclusion that off-pump CABG does not offer any substantial advantages over on-pump CABG except possibly in unusual situations such as, for example, in patients with an extensively calcified (porcelain) aorta, in whom the off-pump technique may result in less manipulation of the aorta, potentially decreasing the risk of aortic emboli or stroke," the authors wrote. "To me, this is an excellent example of the need to rigorously evaluate the benefits in new treatment innovations." said John Spertus (University of Missouri-Kansas City). "On the face of it, off-pump CABG would seem to be an improvement by avoiding the need to manipulate the aorta. However, ROOBY and other similar trials, show less complete revascularization and worse clinical outcomes out to 5 years. From my perspective, this provides compelling evidence that a theoretically better approach to CABG is not as good as traditional bypass surgery. While it is reasonable to perform theoretically better strategies prior to definitive studies, it is also very important to test them against standard approaches and abandon the innovations if they prove not to be associated with significantly better outcomes, as we should now do with off-pump CABG." In an interview, Peterson said that the new paper demonstrates that off-pump CABG is "worse if used in the wrong hands and in the general population." There is no good evidence to support the view that off-pump CABG can reduce neurocognitive side effects that have been linked to off-pump CABG, he said. But he pointed out that there may be a limited role for off-pump CABG in selected patients, such as those with an extensively calcified aorta.