From April 1986 through April 1993, 58 intracoronary stents (41 Wall and 17 Wiktor stents) were implanted for the treatment of saphenous vein graft stenosis in 40 symptomatic patients. The indication was a primary stenosis in 44 and restenosis in 14 procedures. In-hospital complications were subacute stent thrombosis (2%), myocardial infarction (2%), and emergency coronary artery bypass grafting (CABG) (2%). Complications during a mean follow-up period of 42+/-27 months were restenosis (35% by patient, 33% by lesion), myocardial infarction (12%), late bypass grafting (12%), and death (7%). On quantitative coronary angiographic analysis, the mean minimal luminal diameter (and its confidence interval) increased from 1.3 mm (1.1-1.5 mm, preprocedure) to 2.9 mm (2.7-3.1 mm, postprocedure) and 2.2 mm (2.0-2.5 mm, 6 months follow-up, 95% angiographic follow-up). Progression of the underlying coronary artery disease and restenosis were the main reasons for a continual decline of the proportion without cardiac event on a Kaplan-Meier estimate. Restenosis occurred in one-third of cases beyond the first 6 months of follow-up. A relative risk ratio analysis for restenosis, performed on 14 variables, disclosed an increased risk for the following variables: (1) stenting of the proximal, distal or anastomosis part of the vein graft (relative risk 2.41, confidence interval: 1.28-3.59), (2) the implantation of stents < 4.5 mm (2.59, 1.18-4.00), and (3) stenting of a redo-CABG vein graft (2.37, 1.17-3.58). Saphenous vein graft stenting seems to be characterized by excellent immediate clinical and angiographic results; in particular, stent thrombosis is rare.(ABSTRACT TRUNCATED AT 250 WORDS)