LAUSR

Celebrating its 40th anniversary, percutaneous coronary intervention has evolved from balloon angioplasty over bare metal stents to the routine use of drug-eluting stents. Coronary artery stents introduced in the late 1980s marked an important milestone in the development of percutaneous coronary intervention by scaffolding the treated arterial segment and thereby eliminating the risk of abrupt vessel closure due to dissections following balloon angioplasty as well as negative remodelling due to elastic recoil. As a result, percutaneous coronary intervention became a safe and reproducible procedure without the need for standby surgical backup, and emerged as the preferred revascularisation method in most patients with obstructive coronary artery disease. Although current-generation drug-eluting stents are associated with excellent clinical outcomes in terms of restenosis and rates of stent thrombosis of less than 1%, approximately half of clinical adverse events following percutaneous coronary intervention are attributed to device failures rather than progression of atherosclerosis at remote sites. Bioresorbable vascular scaffolds (BVS) that only temporarily scaffold the intervened arterial segment followed by complete biodegradation have been envisioned to overcome long-term failures of metallic stents, with the promising corollaries of restoration of coronary vasomotion, positive vessel remodelling, protective sealing of atherosclerotic plaques, and facilitated access for future bypass anastomoses. The systematic review and meta-analysis by Ziad Ali and colleagues in The Lancet evaluated the 2-year outcomes of the Absorb BVS (Abbott Vascular, Santa Clara, CA, USA) compared with metallic everolimus-eluting stents (EES) in 5583 patients included in seven randomised trials, with individual patient data available from the four industry-sponsored trials with 3389 patients. According to aggregate-level and individual patient data meta-analyses, Absorb BVS were found to be inferior to EES for the device-oriented endpoint, a composite of cardiac mortality, target-vessel myocardial infarction, or ischaemia-driven target-lesion revascularisation. Although there was no difference in mortality, Absorb BVS were associated with a 52% increase in the risk of myocardial infarction and a 40% increase in ischaemia-driven target-lesion revascularisation (both increases were significant). The relative risk of scaffold thrombosis was more than three times higher throughout the entire follow-up period in BVS than in EES (2·3% [73 of 3187] vs 0·7% [16 of 2281]; relative risk [RR] 3·35 [95% CI 1·96–5·72], p<0·0001), with a concerning ten times higher relative risk of very late scaffold thrombosis between 1 year and 2 years (0·8% [24 of 3005] in the BVS group vs 0·1% [two of 2104] in the EES group; RR 9·67 [2·04–45·82], p=0·0042). What does this meta-analysis add to the previously published pooled analysis of individual patient data and other meta-analyses? First, the current findings by Ali and colleagues provide evidence beyond any reasonable doubt that Absorb BVS is inferior to new-generation EES on both efficacy and safety throughout 2 years. Compared with previous analyses published in 2016, follow-up time has accumulated, relative risk increases have become more pronounced, the precision of estimates improved, the power of the meta-analysis increased, and p values— which were at best borderline significant in 2016, providing weak to moderate evidence—now are highly significant, providing strong to overwhelming evidence against the null hypothesis of no difference between BVS and EES. Second, reported results translate into numbers needed to harm to cause one event for up to 2 years when using BVS rather than EES of 47 (95% CI 25–169) for the device-oriented composite endpoint, 45 (26–117) for myocardial infarction, and 61 (31–149) for device thrombosis. It cannot be anticipated that the potential long-term benefits of the current version of Absorb BVS will offset the excess in adverse events reported during the first 2 years. Third, after exclusion of patients with device thrombosis, the relative risk increase in the device-oriented composite endpoint diminished considerably and was no longer significant, suggesting that scaffold thrombosis is the key factor to be targeted in future device iterations. Indeed, rates of very late scaffold thrombosis with Absorb BVS are reminiscent of those observed with early-generation drug-eluting stents. Fourth, the treatment effect against Absorb BVS for the primary device-oriented composite endpoint was consistent across subgroups in 25 of 28 performed analyses. The three remaining analyses on the influence Published Online July 18, 2017 http://dx.doi.org/10.1016/ S0140-6736(17)31871-8