Randomised trials in a highly selected patient population have demonstrated a dramatic reduction in the incidence of in-stent restenosis (ISR) following implantation of sirolimus-eluting (S-E) Cypher coronary stents compared with bare metal stents (BMS). The clinical outcome following implantation of S-E stents for treatment of complex, unselected BMS ISR is less well defined. The aim of this study was to assess the safety and efficacy of S-E coronary stents in the treatment of an unselected population of BMS ISR. All patients who received S-E stents for treatment of BMS ISR from May 1 2002-November 30 2003 at a single institution were entered into a prospectively collected database. In-hospital and long-term outcomes were collected. Sixty patients were identified who received S-E stents for the treatment of ISR. Four patients (6%) had undergone previous brachytherapy and 22% were diabetic. The most common target vessel was the left anterior descending coronary artery (40%), and 6% of lesions were in saphenous vein grafts (SVGs). The mean reference diameter was 2.67+/-0.52 (range 1.75-4.0) mm and the mean lesion length was 16.22+/-11.46 (range 3-68) mm. There were no procedural or in-hospital major adverse cardiac events (MACE). Long-term follow-up was available in 59 patients (98%). The 12-month MACE rate (cardiac death, myocardial infarction or target lesion revascularisation) was 12% with a 7% percutaneous coronary intervention rate and a 7% coronary artery bypass graft rate. There were no cardiac deaths and two non-cardiac deaths. Of the seven patients who had clinical restenosis at 12 months, four had previously failed brachytherapy and three involved SVGs. In conclusion, the use of S-E stents appears safe and efficacious in the treatment of an unselected population of BMS ISR with results comparing favourably with historical controls. Further randomised studies are needed to delineate the optimal management of this high risk group of patients.