Abstract 83: 30-Day Mortality After Carotid Artery Stenting (CAS) in Trial Versus Real-World Settings

Background and Purpose: Clinicians rely on randomized controlled trial (RCT) evidence to select treatments but the risk/benefit profile may differ in the real-world. We compared 30-day mortality after CAS among Medicare, contemporary US registry, and landmark RCT patients. Methods: We linked Medicare data (2000-2009) to CMS’s CAS Database (CAS-D; 2005-2009) and the Society for Vascular Surgery’s Vascular Registry (SVS-VR; 2005-2008) and estimated 30-day mortality among beneficiaries at least 66 years of age. We compared 30-day mortality risk among our Medicare-linked registry patients to those of landmark RCT and post-marketing commitment registries. Results: Average age was lower among RCT patients than among post-marketing registries or Medicare-linked registry patients while symptomatic patients were less common in post-marketing than in Medicare-linked registries or RCTs (Table). SAPPHIRE was the only RCT enrolling high-surgical risk patients but the majority of real-world patients were at high-surgical risk. RCT and post-marketing surveillance registries had minimum provider proficiency requirements while the CAS-D and SVS-VR did not. 30-day mortality risk among Medicare-linked CAS-D (n=22,516) and SVS-VR (n=1,999) was 1.7% (95% CI: 1.5-1.8%) and 1.8% (95% CI: 1.2-2.4%), respectively. 30-day mortality risks among symptomatic Medicare patients were 1.6 to 3.8 times that of RCT CAS patients and mortality risks among RCT patients were lower than among Medicare-linked asymptomatic patients. 30-day mortality risks among registries providing physician training was ~1%, between that of RCTs and Medicare-linked registry patients. Conclusion: Medicare CAS patients have much higher 30-day mortality risks than patients in RCTs or registries with provider training programs because of differences in age, comorbidity burden, and provider proficiency. These findings underscore the importance of evaluating CAS effectiveness in the real-world.