Background In response to the increasing concem with the safety of the drug-eluting stent (DES), the present study aimed to evaluate the long-term safety and efficacy of DES used for a Chinese patient population. Methods All patients, who underwent an index elective percutaneous coronary intervention with an implantation of either DES or bare-metal stent (BMS) in a single institution from April 2004 to December 2006, were included in the analysis. A propensity-score matching technique was applied to adjust and to minimize the impact of confounding factors. Results Overall, there were 1465 patients (20.2%) who had undergone an implantation of only BMS, and 5769 patients (79.8%) of only DES. The propensity-score matching technique set up 1321 pairs of patients for analysis. There were no significant differences between the rates of stent thrombosis (definite and probable) of the two groups (1.06% vs 1.21%, P=-0.8580). Although rates of mortality and myocardial infarction (MI) during the 2-year follow-up period had not differed significantly, rates of death/MI (3.0% vs 4.5%, P=0.0263), target-lesion revascularization (TLR, 3.2% vs 8.5, P=0.0001), target-vessel revascularization (TVR, 5.8% vs 9.5%, P <0.0001) and any revascularization (10.0% vs 13.3%, P=0.0066) were significantly lower for the DES group than for the BMS group. Among the patients in whom devices were implanted for off-label indications, the propensity-score matched rates of stent thrombosis, mortality, MI, and death/MI were not significantly different, while rates of TLR, TVR and any revascularization were significantly lower for the DES group than for the BMS group.Conclusions During the 2 years of follow-up post stenting, DES use is associated with lower rates of death/MI, TLR, TVR and any revascularization, compared with BMS, in propensity-score matched Chinese patient populations. In the setting of off-label usage, DES use is also associated with similar advantages.