Revisiting the Network of Drug-Eluting Stent Trials: Bioresorbable Scaffolds Enter the Arena.

SEE PAGE 1203 T en years ago, during the annual European Society of Cardiology meeting in Barcelona, 2 independent meta-analyses suggested that drug-eluting stents (DES), despite being more effective than bare-metal stents (BMS) in reducing restenosis, increase mortality due to very late thrombosis (1). However, this was no longer confirmed by a plethora of randomized controlled trials (RCTs) and network meta-analyses (NMAs), demonstrating contemporary DES to be not only safe, but even safer than BMS (2). Unquestionably, advancements in DES technology have been instrumental to this newfound perception of safety. The quest for the ideal stent has now culminated with the introduction of bioresorbable scaffolds, in the hope that eliminating inert metal from the coronary arteries will further diminish the onset of late events. When DES and bioresorbable scaffolds are confronted on infrequent outcomes such as stent thrombosis, RCTs are usually underpowered to provide definite estimates of an effect. Standard frequentist meta-analyses may also have limitations, in that they rely on the availability of RCTs for a given comparison, and cannot compare more than 2 devices simultaneously. In a NMA, multiple treatments are compared using both direct comparisons within RCTs and indirect comparisons across RCTs based on a common comparator. As such, NMAs optimize the use of existing data, particularly in areas compounded