The length of a coronary lesion is a significant predictor of restenosis after balloon angioplasty. The influence of lesion length has not comprehensively been assessed after coronary stent placement. This study includes 2,736 consecutive patients with coronary stent placement. Only patients with recent or chronic occlusions before the intervention were excluded. Patients were divided in 2 groups: 573 patients with long lesions (> or = 15 mm) and 2,163 patients with short lesions (< 15 mm). There were no significant differences between the groups with respect to the procedural success rate and incidence of subacute thrombosis. One-year event-free survival was lower in patients with long lesions (73.3% vs 80.0%, p = 0.001). Six-month angiography was performed in 82.5% of the eligible patients. The incidence of binary restenosis (> or = 50% diameter stenosis) was higher in patients with long lesions (36.9% vs 27.9%, p <0.001). Similarly, patients with long lesions presented more late lumen loss than those with short lesions (1.29 +/- 0.89 vs 1.07 +/- 0.77 mm, p <0.001). Multivariate models for both binary restenosis and late lumen loss demonstrated that lesion length was an independent risk factor for restenosis. The risk was further increased by multiple stent placement and overlapping stents that were also independent risk factors of restenosis. Stented segment length did not show any independent effect. Therefore, long lesions represent an independent risk factor for restenosis after coronary stent placement. The results of this study suggest that a possible way to reduce the risk is to cover the lesion with a minimal number of nonoverlapping stents.