Abstract 2811: Four-Year Subgroup Analysis of the NEW-SIRIUS Data: Diabetic versus Non-Diabetic Patients
NEW-SIRIUS is the pooled data analysis of the E-SIRIUS (n=352) and C-SIRIUS (n=100) studies, respectively performed in Europe and Canada. These 2 studies, identical in design, are multicenter randomized, double-blind trials examining the safety and efficacy of the sirolimus-eluting stent (SES) vs. an identical bare-metal stent (BMS) in 452 patients to treat de-novo native long lesions lesions in small coronary arteries. The primary endpoint was in-stent minimum lumen diameter (MLD) at 8-month follow-up. Among the secondary endpoints were major adverse cardiac events (MACE) at 1, 6, 9 and 12 months, and annually up to 5 years post-procedure. The superiority of the SES over the BMS was clearly demonstrated both angiographically and clinically. The binary 8-month in-lesion restenosis rate was 5.1% (vs BMS 44.2%, p < 0.001). At 3 years, the clinical benefit of the SES over the BMS was maintained with a significantly lower incidence of major adverse cardiac events (MACE) (12.4% vs 30.8%, p < 0.001), primarily driven by a significant difference in TLR rates (5.8% vs 26.0%, p < 0.001). A subgroup analysis clearly demonstrated that the SES was also effective in diabetic patients (n = 105; 45 SES, 60 BMS): binary in-lesion restenosis was reduced by 82.2% from 56.1% (BMS) to 10.0% (p < 0.001). At 3 years, the clinical benefit of the SES over the BMS was maintained in diabetic patients, with a significantly lower incidence of major adverse cardiac events (MACE) (17.8% vs 38.3%, p = 0.030), primarily driven by a difference in TLR rates (8.9% vs 31.7%, p = 0.008). Following these results, SES appears to overcome the increased risk of in-stent restenosis associated with diabetes. Event-free survival curves for diabetic patients in both groups were still diverging at 3 years with no indication of a “late catch-up” phenomenon. Four-year clinical follow-up of patients enrolled in E-SIRIUS and C-SIRIUS will be analyzed by July 2006. Results for the overall patient cohort as well as for the diabetic population will be presented.