Background and Objectives:The optimal treatment for in-stent restenosis (ISR) is controversial, although intracoronary radiation therapy (ICRT) has provided the most consistent results to date. This study was designed to assess the early and late angiographic results, and to find independent predictors of recurrent restenosis, following cutting balloon angioplasty (CBA) for ISR. Subjects and Methods:Eighty patients (87 lesions) with first time ISR underwent CBA and systematic follow-up (FU) angiography. A conventional balloon was used before, or after, the CBA, if required. ICRT was used in 18 lesions (21%). A multivariate logistic regression analysis was performed. (Why?) Results:The ISR was focal (n=32, 37%), diffuse or proliferative (n=51, 58%) and occlusive (n=4, 5%). Procedural success was achieved in all 87 lesions (100%). No significant edge dissection occurred. The preand post-procedural diameter stenoses (DS) were 81.5±10.8% and 6.7±6.0%, respectively, and the preand post-procedural MLD (Define MLD?) 0.71±0.44 mm and 2.85±0.32 mm, respectively, with 2.14±0.44 mm of acute gain. A FU angiography was performed in 54 (78%) of the 69 lesions treated with CBA alone. The overall angiographic restenosis rate was 24% (13/54), with 9% (2/22) in the focal ISR and 34% (11/32) in the diffuse or occlusive ISR. The FU DS and MLD were 32.0±23.4% and 2.1±0.7 mm, respectively, with 0.79±0.69 mm of late loss. The length of a restenotic lesion (OR 12.2, 95% CI:1.3115.2, p=0.0286) was an independent predictor of recurrent restenosis. Conclusion:CBA is a simple and efficient first line treatment for ISR, with an acceptable restenosis rate, and the length of a restenotic lesion is an independent predictor of recurrent restenosis. In diffuse or occlusive ISR, more definite treatment modalities, such as ICRT combined with CBA or debulking techniques, might be required to reduce recurrent restenosis. (Korean Circulation J 2003;33 (3):196-204)