The aim of the study was to evaluate the long-term outcomes following selective implantation of drug-eluting stents (DES) in patients at high risk of restenosis versus bare metal stents (BMS) in low-risk patients, according to predefined criteria. Patients who underwent elective percutaneous coronary intervention (PCI) between May 2002 and April 2004 were enrolled in this retrospective, single-centre study. All patients received a BMS while undergoing PCI, unless they fulfilled at least two entry criteria that warranted DES usage. The study endpoints were major adverse cardiac events (MACE), comprising death, myocardial infarction, stent thrombosis (ST), and target vessel revascularisation (TVR), at four years between the DES and BMS groups. A total of 1,250 patients were enrolled in the study, among whom 1,095 (88%) received BMS and the rest received DES. At four years, there was no difference in the cumulative incidence of MACE: death (4.5% in DES vs. 5.8% in BMS, p=0.531), myocardial infarction (2.6% in DES vs. 3.1% in BMS, p=0.722), TVR (9.7% in DES vs. 7.9% in BMS, p=0.461), and ST (1.9% in DES vs. 0.8% in BMS, p=0.183). The event-free survival rate at four years was similar in the two groups (87.1% in DES vs. 86.1% in BMS; p=0.741). In elective PCI, a strategy of selective use of DES in patients at high risk of restenosis based on predefined criteria confers the same favourable long-term clinical outcomes as BMS in low-risk patients.