OBJECTIVE: To investigate the safety and efficacy of endoluminal vessel reconstruction by intracoronary stenting in patients with single or multivessel disease who have been turned down for surgical revascularisation. DESIGN: An observational, single centre trial of consecutive patients being followed up clinically. SETTING: Tertiary referral centre. PATIENTS: 43 patients treated between 1995 and 1996 with a clinical follow up period of at least six months. MAIN OUTCOME MEASURES: Acute procedural success following intervention (defined as a residual stenosis of < 30% with normal antegrade flow and without acute clinical complications); major adverse clinical events (death, myocardial infarction, repeat intervention, clinical restenosis); and functional status during follow up. RESULTS: Acute procedural success was achieved in 39 patients. Failures resulted from a refractory no reflow phenomenon with Q wave myocardial infarction (two patients), and from subacute stent thrombosis (one patient) and saphenous vein graft rupture at stent site (one patient), both manifesting by non-Q wave myocardial infarction. During a follow up period of 17 (8) months (mean (SD)), three of these patients died. Clinical restenosis was treated by repeat percutaneous intervention in 11 patients, while 28 patients remained asymptomatic. On a Kaplan-Meier estimate, the proportion of patients without adverse clinical events at two years was 65%. CONCLUSIONS: Percutaneous reconstruction of diseased coronary arteries in patients with single or multivessel disease turned down for surgical revascularisation may be considered as a valuable treatment option. On follow up, more events seem to be encountered than after elective de novo stenting of discrete coronary lesions. This is probably related to higher clinical restenosis rates.