Mid- and Long-Term Outcome Comparisons of Everolimus-Eluting Bioresorbable Scaffolds Versus Everolimus-Eluting Metallic Stents: A Systematic Review and Meta-analysis

Background Percutaneous coronary interventions to implant bioresorbable vascular scaffolds (BVSs) were designed to reduce the late thrombotic events that occur with metallic stents. Purpose To estimate the incidence of scaffold thrombosis after BVS implantation and compare everolimus-eluting BVSs with everolimus-eluting metallic stents (EESs) in terms of safety and efficacy at mid- and long-term follow-up in adults who had a percutaneous coronary intervention. Data sources PubMed, EMBASE, the Cochrane Library, conference proceedings, and relevant Web sites from inception until 20 May 2017, without language restriction. Study selection 7 randomized trials and 38 observational studies (each with a minimum of 6 months and 100 patient-years of follow-up) in adults with coronary artery disease who had a BVS or an EES and reported scaffold or stent thrombosis (main outcome) or other secondary outcomes (such as death, myocardial infarction, or revascularization). Data extraction 2 reviewers independently extracted study data, rated study quality, and assessed strength of evidence. Data synthesis The pooled incidence of definite or probable scaffold thrombosis after BVS implantation was 1.8% (95% CI, 1.5% to 2.2%) at a median follow-up of 1 year (41 studies, 21 884 patients) and 0.8% (CI, 0.5% to 1.3%) beyond 1 year (14 studies, 4688 patients). Seven trials involving 5578 patients that directly compared BVSs with EESs showed an increased risk for definite or probable scaffold thrombosis (odds ratio [OR], 3.40 [CI, 2.01 to 5.76]) with BVSs at a median follow-up of 25 months. Increased risks were present at early (prominently subacute), late, and very late stages, and odds beyond 1 year were almost double those seen within 1 year. Bioresorbably vascular scaffolds increased risks for myocardial infarction (OR, 1.63 [CI, 1.26 to 2.10]), target lesion revascularization (OR, 1.31 [CI, 1.03 to 1.67]), and target lesion failure (OR, 1.37 [CI, 1.12 to 1.66]); the odds for these 3 end points also increased over time. The incidences of all-cause, cardiac, and noncardiac death and of target vessel and any revascularization did not differ. Limitation Quality of observational studies was unclear, and some data were unpublished. Conclusion Compared with EESs, BVSs increased the risks for scaffold thrombosis and other thrombotic events at mid- and long-term follow-up, and risks increased over time. Primary funding source National Natural Science Foundation of China.