Objectives: We assessed outcomes of patients undergoing drug-eluting stent (DES) vs. bare metal stent (BMS) implantation for complex lesions excluded from pivotal clinical trials of DES. Background: Although DES improve target vessel revascularization (TVR) and major adverse cardiovascular events (MACE) compared to BMS in randomized trials, data on safety and efficacy of DES in complex lesions are insufficient. Methods: In a single-center registry of 1,354 patients who underwent stent implantation for complex lesions between July 2001 and December 2005, we compared the incidence of death, death or myocardial infarction (MI), stent thrombosis [definite or probable by the Academic Research Consortium (ARC) criteria], TVR, and MACE between patients who received DES (n = 483) versus those who received BMS (n = 871). Mean duration of follow-up was 494 versus 838 days in DES and BMS groups, respectively. Results: Clinical outcomes in DES versus BMS groups were as follows: death 5.2% versus 11.5% (log-rank P = 0.042); death/MI 11.2% versus 16.7% (P = 0.47), stent thrombosis 2.9% versus 2.6% (P = 0.61), TVR 6.6 versus 18.5% (P < 0.0001), MACE 14.9% versus 29.7% (P = 0.0002), respectively. After adjustment for baseline differences, DES implantation was associated with lower TVR (adjusted hazards ratio HR = 0.38, 95% CI 0.26-0.56, P < 0.0001) and MACE (HR = 0.56, CI 0.42-0.74, P < 0.0001) without significant impact on other outcomes. In 933 patients who underwent DES (n = 483) or BMS (n = 450) implantation in the year 2003 or later, DES implantation similarly lowered TVR and MACE without affecting other outcomes. Conclusions: Our findings support the safety and efficacy of DES in patient subsets excluded from pivotal randomized clinical trials of DES.