Comparison of an everolimus-eluting bioresorbable scaffold with an everolimus-eluting metallic stent for the treatment of coronary artery stenosis (ABSORB II): a 3 year, randomised, controlled, single-blind, multicentre clinical trial

Background No medium-term data are available on the random comparison between everolimus-eluting bioresorbable vascular scaffolds and everolimus-eluting metallic stents. The study aims to demonstrate two mechanistic properties of the bioresorbable scaffold: increase in luminal dimensions as a result of recovered vasomotion of the scaffolded vessel. Methods The ABSORB II trial is a prospective, randomised, active-controlled, single-blind, parallel two-group, multicentre clinical trial. We enrolled eligible patients aged 18–85 years with evidence of myocardial ischaemia and one or two de-novo native lesions in different epicardial vessels. We randomly assigned patients (2:1) to receive treatment with an everolimus-eluting bioresorbable scaffold (Absorb; Abbott Vascular, Santa Clara, CA, USA) or treatment with an everolimus-eluting metallic stent (Xience; Abbott Vascular, Santa Clara, CA, USA). Randomisation was stratified by diabetes status and number of planned target lesions. At 3 year follow-up, the primary endpoint was superiority of the Absorb bioresorbable scaffold versus the Xience metallic stent in angiographic vasomotor reactivity after administration of intracoronary nitrate. The co-primary endpoint is the non-inferiority of angiographic late luminal loss. For the endpoint of vasomotion, the comparison was tested using a two-sided t test. For the endpoint of late luminal loss, non-inferiority was tested using a one-sided asymptotic test, against a non-inferiority margin of 0·14 mm. The trial is registered at ClinicalTrials.gov, number NCT01425281. Findings Between Nov 28, 2011, and June 4, 2013, we enrolled 501 patients and randomly assigned them to the Absorb group (335 patients, 364 lesions) or the Xience group (166 patients, 182 lesions). The vasomotor reactivity at 3 years was not statistically different (Absorb group 0·047 mm [SD 0·109] vs Xience group 0·056 mm [0·117]; psuperiority=0·49), whereas the late luminal loss was larger in the Absorb group than in the Xience group (0·37 mm [0·45] vs 0·25 mm [0·25]; pnon-inferiority=0·78). This difference in luminal dimension was confirmed by intravascular ultrasound assessment of the minimum lumen area (4·32 mm2 [SD 1·48] vs 5·38 mm2 [1·51]; p<0·0001). The secondary endpoints of patient-oriented composite endpoint, Seattle Angina Questionnaire score, and exercise testing were not statistically different in both groups. However, a device-oriented composite endpoint was significantly different between the Absorb group and the Xience group (10% vs 5%, hazard ratio 2·17 [95% CI 1·01–4·70]; log-rank test p=0·0425), mainly driven by target vessel myocardial infarction (6% vs 1%; p=0·0108), including peri-procedural myocardial infarction (4% vs 1%; p=0·16). Interpretation The trial did not meet its co-primary endpoints of superior vasomotor reactivity and non-inferior late luminal loss for the Absorb bioresorbable scaffold with respect to the metallic stent, which was found to have significantly lower late luminal loss than the Absorb scaffold. A higher rate of device-oriented composite endpoint due to target vessel myocardial infarction, including peri-procedural myocardial infarction, was observed in the Absorb group. The patient-oriented composite endpoint, anginal status, and exercise testing, were not statistically different between both devices at 3 years. Future studies should investigate the clinical impact of accurate intravascular imaging in sizing the device and in optimising the scaffold implantation. The benefit and need for prolonged dual antiplatelet therapy after bioresorbable scaffold implantation could also become a topic for future clinical research. Funding Abbott Vascular.