The authors of this study hypothesized that percutaneous transluminal coronary angioplasty of a stenosis at the ostium of a branch vessel, whether isolated or associated with a bifurcation stenosis, was associated with reduced procedural success and increased in-hospital complications. One hundred six patients with 119 ostial branch stenoses were compared with 1,168 patients who underwent angioplasty of nonostial branch stenoses. An ostial branch stenosis was defined as a stenosis in the proximal 3 mm of a major branch vessel (diagonal [n = 58], posterior descending [n = 21], obtuse marginal [n = 34] and intermediate [n = 6]). The ostial branch stenosis was isolated in 61% of the patients and associated with a bifurcation stenosis in 39%. Despite a balloon to artery ratio of 1.05:1, angiographic success was 74% of ostial branch stenoses versus 91% of nonostial stenoses (p less than 0.01). Furthermore, angioplasty of ostial branch stenoses resulted in a complication rate of 13 versus 5% for angioplasty of nonostial branch stenoses (p less than 0.01). Therefore, angioplasty of ostial branch stenoses results in decreased procedural success and significant residual stenosis despite adequate balloon sizing, suggesting arterial elastic recoil and a significant increase in complications.