Predictors of repeat revascularization after nonemergent, first percutaneous coronary intervention in the community

Background We sought to determine the incidence of and risk factors for repeat revascularization after nonemergent, first percutaneous coronary intervention (PCI) performed in contemporary community practice. Methods We analyzed a prospective registry of consecutive patients undergoing isolated PCI in the state of Washington. Multivariate Cox regression analysis was used to determine predictors of repeat revascularization (by PCI or bypass surgery) within 1 year after first PCI. Results Between January 1, 1999, and December 31, 1999, there were 3571 nonemergent first PCIs, 87.7% of which involved stent placement. Repeat revascularization occurred in 577 (16.2%) patients. Repeat revascularization was predicted by multivessel disease (hazard ratio [HR] 1.36, 95% CI 1.12–1.66), stable versus no angina (HR 1.27, 95% CI 1.03–1.57), and maximum stent length used (per 1 mm longer: HR 1.01, 95% CI 1.002–1.02), while prior myocardial infarction (HR 0.77, 95% CI 0.62–0.96) and creatinine >1.2 mg/dL (HR 0.74, 95% CI 0.56–0.98) were associated with lower risk of repeat revascularization. Diabetes was a significant predictor only when the outcome was limited to revascularization by coronary artery bypass surgery (HR 1.52, 95% CI 1.03–2.23). Although glycoprotein IIb/IIIa inhibitor use was a significant univariate predictor of freedom from early repeat revascularization (within 60 days after first PCI), after controlling for potential confounders, it was no longer significant. Conclusions In this contemporary, community-based registry of patients undergoing nonemergent first PCI, clinical practice and outcomes are consistent with evidence from clinical trials and previous controlled studies. Results from controlled studies may reasonably be extrapolated to such a community setting.