Prospective follow-up at 2 years was obtained for 98.7% of the pooled 1,510 patients enrolled in SIRIUS, E-SIRIUS and C-SIRIUS, 3 randomized controlled trials that compared sirolimus-eluting stents (SESs) with bare metal stents (BMSs) to treat long stenoses in small coronary arteries. By 720 days, clinically driven target lesion revascularizations were performed in 5.7% of patients with SESs versus 22.6% of patients with BMSs (risk ratio 0.25, 95% confidence interval 0.18 to 0.35, p <0.001). Of these, late target lesion revascularization (from 271 to 720 days) was performed in 12 patients who received SESs (1.6%) compared with 37 patients with BMSs (4.9%) (risk ratio 0.32, 0.17 to 0.61, p <0.001). Stent thromboses occurred in 7 of 758 patients with SESs (0.9%, 4 subacute, 3 late) and 5 of 752 patients with BMSs (0.7%, 1 subacute, 4 late) (risk ratio 1.39, 95% confidence interval 0.44 to 4.36, p = 0.774). The Kaplan-Meier estimate of freedom from major cardiac adverse events was 89.3% for patients with SESs versus 73.4% for patients with BMSs (p <0.001). This analysis demonstrates the sustained efficacy and safety of sirolimus-eluting stents at 2 years, characterized by a persistent significant benefit in freedom from repeat revascularization compared with BMSs and a low risk of late stent thrombosis, not different from BMSs.