Research in context
Evidence before this study
We searched PubMed, several websites (www.tctmd.com, www.clinicaltrials.gov, www.clinicaltrialresults.org, and www.acc.org) and the proceedings of major cardiology meetings within the past 5 years (Jan 1, 2010–Oct 1, 2015) to identify randomised trials of bioresorbable vascular scaffolds (BVS) versus metallic drug-eluting stents. We did not use any language restrictions in our search. We identified six candidate trials: ABSORB II, ABSORB Japan, ABSORB China, ABSORB III, EVERBIO II, and TROFI II. Individual review of these trials showed non-significantly different 1-year outcomes between the everolimus-eluting Absorb BVS and the Xience cobalt-chromium everolimus-eluting stent (CoCr-EES) for most clinical endpoints, with some exceptions (lower all-cause mortality with BVS vs CoCr-EES in ABSORB China, and greater subacute device thrombosis between 1 and 30 days with BVS than with CoCr-EES in ABSORB III). However, only two of these trials were powered for clinical endpoints (ABSORB III and ABSORB Japan) and none were sufficiently large to detect differences in low-frequency endpoints between BVS and CoCr-EES.
Added value of this study
This patient-level, pooled meta-analysis of four trials in 3389 randomly assigned patients provides substantially greater power to describe the safety and effectiveness profile of BVS versus CoCr-EES than any of the individual studies alone. Our analysis shows similar results for BVS and CoCr-EES for the patient-oriented composite endpoint and the device-oriented composite endpoint at 1 year, which are arguably the most important overall measures of patient-related and device-related clinical outcomes. These findings were similar if the two additional randomised trials with less than 1 year follow-up (EVERBIO II and TROFI II) were added. An increase in target vessel-related myocardial infarction with BVS was observed, which might be attributable to non-significant increases in peri-procedural myocardial infarction and device thrombosis, although no differences between the groups were noted in cardiac or all-cause mortality, all myocardial infarction, or revascularisation measures of efficacy.
Implications of all the available evidence
Despite BVS being a first-generation technology for which most users are still learning the optimum implantation technique, and despite the fact that in these trials BVS was compared against CoCr-EES—the metallic drug-eluting stent with the lowest rate of stent thrombosis—the aggregate of available evidence supports the safety and effectiveness of BVS at 1 year for treatment of patients with stable coronary artery disease and stabilised acute coronary syndromes. Attention to appropriate device sizing, adequate lesion preparation, and routine high-pressure post-dilatation might further improve the 30-day and 1-year outcomes of BVS. Long-term results from the present and additional ongoing large-scale trials are needed to ascertain whether or not the novel properties of BVS result in improved late outcomes in patients with coronary artery disease undergoing percutaneous coronary intervention.