Three-year outcomes associated with embolic protection in saphenous vein graft intervention: results in 49 325 senior patients in the Medicare-linked National Cardiovascular Data Registry CathPCI Registry.

BACKGROUND Information is limited on contemporary use and outcomes of embolic protection devices (EPDs) in saphenous vein graft interventions. METHODS AND RESULTS We formed a longitudinal cohort (2005-2009; n=49 325) by linking National Cardiovascular Data Registry CathPCI Registry to Medicare claims to examine the association between EPD use and both procedural and long-term outcomes among seniors (65+ years), adjusting for clinical factors using propensity and instrumental variable methodologies. Prespecified high-risk subgroups included acute coronary syndrome and de novo or graft body lesions. EPDs were used in 21.2% of saphenous vein grafts (median age, 75; 23% women) and were more common in acute coronary syndrome (versus non-acute coronary syndrome; 22% versus 19%), de novo (versus restenotic; 22% versus 14%), and graft body lesions (versus aortic and distal anastomosis; 24% versus 20% versus 8%, respectively). EPDs were associated with a slightly higher incidence of procedural complications, including no reflow (3.9% versus 2.8%; P<0.001), vessel dissection (1.3% versus 1.1%; P=0.05), perforation (0.7% versus 0.4%; P=0.001), and periprocedural myocardial infarction (2.8% versus 1.8%; P<0.001). By 3 years, death, myocardial infarction, and repeat revascularization occurred in 25%, 15%, and 30% of cases, respectively. EPD use was associated with a similar adjusted risk of death (propensity score-matched hazard ratio, 0.96; 95% confidence interval, 0.91-1.02), myocardial infarction (propensity score-matched hazard ratio, 1.00; 95% confidence interval, 0.93-1.09), and repeat revascularization (propensity score-matched hazard ratio, 1.02; 95% confidence interval, 0.96-1.08) in the overall cohort and high-risk subgroups. CONCLUSIONS In this contemporary cohort, EPDs were used more commonly among patients with high-risk clinical indications, yet there was no evidence of improved acute- or long-term outcomes. Further prospective studies are needed to support routine EPD use.