Long-Term Safety and Efficacy of Paclitaxel-Eluting Stents: Final 5-Year Analysis From the TAXUS Clinical Trial Program

Abstract

These studies sought to evaluate the clinical outcomes of the slow-release Taxus paclitaxel-eluting stent (PES) versus an otherwise identical bare-metal stent (BMS). Prior studies were not individually powered to generate reliable estimates of low-frequency safety endpoints or to characterize the long-term safety and efficacy profile of PES. The completed 5-year databases from the prospective, randomized, double-blind TAXUS I, II, IV, and V trials were pooled for a patient-level analysis. The study population comprised 2,797 randomized patients (1,400 PES and 1,397 BMS). At the end of the 5-year study period, PES compared with BMS significantly reduced the rate of ischemia-driven target lesion revascularization (12.3% vs. 21.0%, p < 0.0001), with consistent reductions across high-risk subgroups and in patients with and without routine angiographic follow-up. There were no significant differences between the stent types in the 1-year or cumulative 5-year rates of death or myocardial infarction (MI). However, cardiac death or MI between 1 and 5 years was increased with PES (6.7% vs. 4.5%, p = 0.01), as was stent thrombosis (protocol definition: 0.9% vs. 0.2%, p = 0.007; ARC definition: 1.4% vs. 0.9%, p = 0.18). In this pooled patient-level analysis from the prospective, randomized, double-blind TAXUS trials, PES compared with BMS resulted in a durable 47% reduction in the 5-year rate of ischemia-driven target lesion revascularization in simple and complex lesions, with nonsignificant differences in the cumulative 5-year rates of death or MI. Between 1 and 5 years, however, the rates of cardiac death or MI and protocol-defined stent thrombosis were increased with PES.