To determine the results of coronary angioplasty for a first restenosis, the clinical, anatomic, and procedural data of 400 consecutive patients were compared with the data of 507 consecutive patients undergoing a first angioplasty. After angioplasty for restenosis, emergency redilatation had to be performed in only 0.7% of the patients versus 3.1% of the control group (p = 0.02); nevertheless, the major in-hospital event (death, myocardial infarction, emergency coronary surgery, cerebrovascular accident) rate for patients was only slightly lower (3.3% vs 4.2%, p = NS). During the 6-month follow-up period, there were no cardiac deaths and only two myocardial infarctions in the study group, but recurrent ischemia was more frequent (37% vs 31%, p = 0.05) and resulted in considerably more elective coronary surgery (16% vs 2.6%, p = 0.001). In the study group, stepwise discriminant analysis revealed four variables significantly related to the occurrence of a second restenosis: time interval between first and second angioplasty, male gender, severity of angina, and complexity of the restenotic lesions. However, their individual predictive power was low. In conclusion, compared with angioplasty for primary lesions, angioplasty for restenosis was associated with fewer periprocedural complications and, after a 6-month follow-up, serious cardiac events were almost nonexistent but recurrent ischemia was more frequent.