Weighing the potential late benefits versus early hazard associated with bioresorbable vascular scaffolds in percutaneous coronary interventions: a Markov decision analytic model.

Abstract

Use of poly-L-lactic acid-based bioresorbable scaffolds (BRS) has been associated with increased risk of device thrombosis during the first 3 years after implantation as compared to metallic everolimus-eluting stents (EES). The long-term performance of BRS relative to EES remains unknown. We used a Markov decision analysis model to evaluate the effectiveness of BRS vs. EES over a lifetime horizon. In addition to one-way sensitivity analyses of key variables, we evaluated the impact of optimal implantation technique and limiting procedures to larger vessels (>2.6 mm in diameter) on model results. Assuming no risk of target lesion revascularization for BRS after 3 years, we found a small increment in quality-adjusted life expectancy (QALE) of 0.02 with the use of BRS relative to EES, with benefit being observed after 21.8 years. Optimal implantation technique and limiting to larger vessels resulted in larger gains in QALE (0.08 and 0.06, respectively) with BRS and shorter times to equipoise (6.7 and 8.3 years, respectively). Model results were highly sensitive to variations in the relative risk of stent thrombosis (BRS vs. EES). Based on currently available data, it would take approximately 21.8 years for the presumed late benefits of current BRS relative to EES to overcome the early hazard associated with their use under favorable assumptions. Optimal implantation technique and limiting procedures to larger vessels improved BRS performance and reduced time to equipoise. Eliminating the higher BRS thrombosis risk is necessary in developing future generations of BRS as an acceptable alternative to EES.