In patients with diabetes mellitus, outcome after drug-eluting stent (DES) versus bare metal stent (BMS) implantation remains under investigation; although lower reintervention rates were reported, incidence of death and myocardial infarction (MI) during follow-up is not completely characterized. Thus, we performed a meta-analysis of available randomized studies evaluating follow-up events of DESs versus BMSs in patients with diabetes mellitus. Randomized trials reporting outcome of DES versus BMS in diabetic patients with a follow-up > or =6 months were included. Outcomes analyzed were (1) death, (2) MI, (3) in-stent restenosis (ISR) and target lesion revascularization (TLR), and (4) stent thrombosis. Data were extracted by 2 independent reviewers. A total of 9 trials, including 1,141 patients, were found. ISR occurred in 8% of patients with DESs versus 41% of those with BMSs (odds ratio [OR] 0.13, 95 confidence interval [CI] 0.09 to 0.20, p <0.00001) and TLR in 8% versus 27% (OR 0.23, 95% CI 0.16 to 0.33, p <0.00001). There was no difference in the incidence of stent thrombosis (1.1% vs 1.2%, p = 0.98) or death (2.4% vs 2.3%, p = 0.91). MI occurred in 3.5% of patients with DESs versus 7.2% of those with BMSs (52% risk decrease, p = 0.02). Decrease of ISR with DESs was observed in noninsulin-treated (OR 0.17, 95% CI 0.11 to 0.26, p <0.00001) and insulin-treated (OR 0.22, 95% CI 0.13 to 0.37, p <0.00001) patients. In conclusion, diabetic patients receiving DESs have lower risk of ISR and TLR versus those treated with BMSs; use of DESs in patients with diabetes mellitus significantly decreases the incidence of MI during follow-up, without affecting mortality or stent thrombosis.