Objectives We undertook a meta-analysis to assess outcomes for drug-eluting stents (DES) and bare-metal stents (BMS) in percutaneous coronary intervention for unprotected left main coronary artery (ULMCA) stenosis. Background Uncertainty exists regarding the relative performance of DES versus BMS in percutaneous coronary intervention for unprotected left main coronary stenosis. Methods Of a total of 838 studies, 44 met inclusion criteria (n = 10,342). The co-primary end points were mortality, myocardial infarction (MI), target vessel/lesion revascularization (TVR/TLR), and major adverse cardiac events (MACE: mortality, MI, TVR/TLR). Results Event rates for DES and BMS were calculated at 6 to 12 months, at 2 years, and at 3 years. Crude event rates at 3 years were mortality (8.8% and 12.7%), MI (4.0% and 3.4%), TVR/TLR (8.0% and 16.4%), and MACE (21.4% and 31.6%). Nine studies were included in a comparative analysis (n = 5,081). At 6 to 12 months the adjusted odds ratio (OR) for DES versus BMS were: mortality 0.94 (95% confidence interval [CI]: 0.06 to 15.48; p = 0.97), MI 0.64 (95% CI: 0.19 to 2.17; p = 0.47), TVR/TLR 0.10 (95% CI: 0.01 to 0.84; p = 0.01), and MACE 0.34 (95% CI: 0.15 to 0.78; p = 0.01). At 2 years, the OR for DES versus BMS were: mortality 0.42 (95% CI: 0.28 to 0.62; p < 0.01), MI 0.16 (95% CI: 0.01 to 3.53; p = 0.13), and MACE 0.31 (95% CI: 0.15 to 0.66; p < 0.01). At 3 years, the OR for DES versus BMS were: mortality 0.70 (95% CI: 0.53 to 0.92; p = 0.01), MI 0.49 (95% CI: 0.26 to 0.92; p = 0.03), TVR/TLR 0.46 (95% CI: 0.30 to 0.69; p < 0.01), and MACE 0.78 (95% CI: 0.57 to 1.07; p = 0.12). Conclusions Our meta-analysis suggests that DES is associated with favorable outcomes for mortality, MI, TVR/TLR, and MACE as compared to BMS in percutaneous coronary intervention for unprotected left main coronary artery stenosis. (J Am Coll Cardiol Intv 2010;3:602-11) (C) 2010 by the American College of Cardiology Foundation