Background Patients with prior coronary artery bypass graft (CABG) have a poor outcome after acute myocardial infarction (AMI).Little is known about the treatment strategy and outcome of percutaneous coronary intervention (PCI) in these patients.The purpose of this study was to investigate the impact of graft versus native artery PCI on the outcomes of prior CABG patients with AMI.Methods Between September 2005 and October 2011,a total of 140 consecutive patients with previous CABG undergoing PCI for the treatment of AMI were included.Clinical/procedural characteristics and long-term clinical outcomes were compared between graft and native artery PCI patients.Results The mean time interval to prior CABG was (5.6±4.2) years.Thirty patients received graft PCI,success rate being 90％.One hundred and ten patients received native artery PCI,success rate being 90.7％ (P ＞0.05).There were no significant differences in the basic characteristics between the two groups.All patients received drug eluting stents (DESs).Three patients died during hospitalization in the graft-PCI group (10％ vs.native PCI 0,P ＜0.05).After a median followup of two years,major adverse cardiac events (MACE) (myocardial infarction,target vessel revascularization,total death) were 20％ with no significant difference between the two groups.Cox regression analysis showed that both diabetes mellitus (DM,HR 3.57,95％ CI 1.03-5.75,P ＜0.05) and primary PCI (HR 5.932,95％ Cl 1.91-18.4,P ＜0.05) were independent predictors of MACE.Conclusions More patients with prior CABG underwent native artery PCI for AMI.PCI to culprit graft vessels had higher in-hospital mortality.DM and primary PCI,but not graft PCI,were predictors for adverse long-term outcome.