Stent thrombosis occurs infrequently after drug-eluting stent (DES) placement, but concerns have surfaced that identification of stent thrombosis based solely on angiography may underestimate the true incidence of this complication. The purpose of this study was to compare rates of stent thrombosis using Academic Research Consortium (ARC) definitions. From April 2002 to February 2005, bare-metal stents (BMSs) and DESs were placed at Wake Forest University Baptist Medical Center. Rates of ARC definite, probable, and possible stent thrombosis were evaluated. Overall, definite stent thrombosis occurred in 1.4% of BMSs and 1.1% of DESs (p = 0.47); probable stent thrombosis, in 1.0% of BMSs and 0.9% of DESs (p = 0.81); definite plus probable stent thrombosis, in 2.3% of BMSs and 1.9% of DESs (p = 0.48); and possible stent thrombosis, in 6.8% of BMSs and 3.5% of DESs (p <0.001). Rates of definite late (31 days to 1 year) and very late (>1 year) stent thrombosis were 0.9% for BMSs and 0.06% for DESs (p <0.001) and 0.16% for BMSs and 0.56% for DESs, respectively (p = 0.13). Rates of definite plus probable and late and very late stent thrombosis were 1.1% for BMSs and 0.19% for DESs (p <0.001) and 0.40% for BMSs and 0.68% for DESs, respectively (p = 0.31). Rates of possible stent thrombosis increased significantly for each thrombosis interval. In conclusion, rates of stent thrombosis using the proposed ARC definition differed by stent type, definition of stent thrombosis, and thrombosis interval. Use of the ARC probable stent thrombosis paralleled rates of definite stent thrombosis alone, but partially masked the occurrence of late (>1 year) stent thrombosis in DESs compared with BMSs.