1-year outcomes with the Absorb bioresorbable scaffold in patients with coronary artery disease: a patient-level, pooled meta-analysis

Summary Background Compared with metallic drug-eluting stents, bioresorbable vascular scaffolds (BVS) offer the potential to improve long-term outcomes of percutaneous coronary intervention. Whether or not these devices are as safe and effective as drug-eluting stents within the first year after implantation is unknown. Methods We did a patient-level, pooled meta-analysis of four randomised trials in which 3389 patients with stable coronary artery disease or a stabilised acute coronary syndrome were enrolled at 301 academic and medical centres in North America, Europe, and the Asia-Pacific region. These patients were randomly assigned to the everolimus-eluting Absorb BVS (n=2164) or the Xience cobalt-chromium everolimus-eluting stent (CoCr-EES; n=1225). The primary endpoints were the 1-year relative rates of the patient-oriented composite endpoint (all-cause mortality, all myocardial infarction, or all revascularisation) and the device-oriented composite endpoint of target lesion failure (cardiac mortality, target vessel-related myocardial infarction, or ischaemia-driven target lesion revascularisation). All analyses were by intention to treat. The four randomised trials included in our meta-analysis are all registered with ClinicalTrials.gov , numbers NCT01751906 , NCT01844284 , NCT01923740 , and NCT01425281. Findings The summary treatment effect for the 1-year relative rates of the patient-oriented composite endpoint did not differ significantly different between BVS and CoCr-EES (relative risk [RR] 1·09 [0·89–1·34], p=0·38). Similarly, the 1-year relative rates of the device-oriented composite endpoint did not differ between the groups (RR 1·22 [95% CI 0·91–1·64], p=0·17). Target vessel-related myocardial infarction was increased with BVS compared with CoCr-EES (RR 1·45 [95% CI 1·02–2·07], p=0·04), due in part to non-significant increases in peri-procedural myocardial infarction and device thrombosis with BVS (RR 2·09 [0·92–4·75], p=0·08). The relative rates of all-cause and cardiac mortality, all myocardial infarction, ischaemia-driven target lesion revascularisation, and all revascularisation did not differ between BVS and CoCr-EES. Results were similar after multivariable adjustment for baseline imbalances, and were consistent across most subgroups and in sensitivity analysis when two additional randomised trials with less than 1 year of follow-up were included. Interpretation In this meta-analysis, BVS did not lead to different rates of composite pa