Percutaneous intervention for the first episode of in-stent restenosis was performed in 177 patients 5.4 +/- 0.3 months after native coronary stent implantation. Medical records were reviewed and patients contacted 13.3 +/- 1.2 months after in-stent intervention to ascertain the subsequent clinical course. The effects of demographic, procedural, and angiographic variables on clinical outcomes were determined. At 2 years, Kaplan-Meier estimated survival was 93 +/- 3% and freedom from death, myocardial infarction, and a third target artery revascularization (TAR) was 67 +/- 4%. The actuarial frequency of a third TAR was 26 +/- 4% at 1 year. Stratification of outcomes according to timing of in-stent intervention revealed an approximate twofold higher frequency of adverse events among patients with early (</=3 months) in-stent restenosis. Advanced age (p = 0.019), prior coronary bypass (p = 0. 017), and early in-stent intervention (p = 0.006) independently predicted increased mortality at follow-up. Systemic hypertension (p = 0.004), diabetes mellitus (p = 0.044), and early in-stent intervention (p <0.0001) independently predicted a third TAR. These variables (p = 0.007, p = 0.027, and p <0.0001, respectively) also independently predicted a composite end point consisting of death, myocardial infarction, and a third TAR. No angiographic variable predicted outcome after in-stent intervention. Thus, long-term outcome following in-stent intervention was favorable. Early in-stent intervention, advanced age, hypertension, and diabetes predicted adverse outcomes.