Clinical and angiographic outcomes in diabetic patients following single or multivessel stenting in the COSTAR II randomized trial

BACKGROUND: Percutaneous coronary intervention (PCI) is associated with increased clinical and angiographic restenosis in diabetic patients. Stent-based elution of paclitaxel from a biostable polymer reduces restenosis and major adverse cardiovascular events (MACE) when compared with bare-metal stent deployment. The safety and efficacy of paclitaxel elution from a bioresorbable polymer has not been studied in diabetic patients. METHODS: Patients (n = 1700) with single- or multivessel coronary disease were randomized (3:2) to receive the CoStar® or Taxus® stent. All patients had glycolated hemoglobin (HbA1c) obtained at enrollment. RESULTS: Medically-treated diabetes was present in 469 patients (117 insulin-treated) and 77 patients had elevated HbA1c > 6.5% in the absence of previously diagnosed diabetes. MACE were increased in diabetics and were greatest in those requiring insulin. Elevated HbA1c, in the absence of diagnosed diabetes, was not associated with adverse outcomes. MACE (8 months) in the diabetic cohort trended lower with Taxus versus CoStar (10.9 vs. 14.4%, respectively; p = 0.271) due to a reduction in target vessel revascularization. Late lumen loss in-segment (9 months) was reduced by Taxus compared to CoStar (0.20 vs. 0.52 mm, respectively; p < 0.05). CONCLUSION: Diabetes is associated with adverse outcomes following stent deployment. Taxus stents improved angiographic outcomes with a trend toward improved clinical outcomes when compared with CoStar stents following PCI in diabetic patients. As a measure of preprocedural glycemic control, the HbA1c level was weakly correlated with outcomes.