The cost-effectiveness of guideline-driven use of drug-eluting stents: propensity-score matched analysis of a seven-year multicentre experience.

Background: In routine clinical practice, the implantation of a drug-eluting stent (DES) versus a bare metal stent (BMS) for percutaneous coronary intervention (PCI) has been guided by criteria for appropriate use. The cost-effectiveness (CE) of adopting these guidelines, however, is not clear, and was investigated from the perspective of the Australian healthcare payer. Methods and results: Baseline and 12-month follow-up data of 12,710 PCI patients enrolled in the Melbourne Interventional Group (MIG) registry between 2004 and 2011 were analysed. Costs inputs were derived from a clinical costing database and published sources. Propensity-score-matching was performed for DES and BMS groups within sub-groups. Incremental cost-effectiveness ratios (ICERs) were evaluated for all patients, and sub-groups of patients with '0', 1, 2, or ≥3 indications for a DES. The incremental cost per target vessel revascularization avoided for the overall population was $24,683, and for patients with 0, 1, and 2 indications for a DES was $44,635, $33,335, and $23,788, respectively. However, for those with >3 indications, DES compared with BMS was associated with cost savings. At willingness to pay thresholds of $45,000-$75,000, the probability of cost-effectiveness of DES for the overall cohort was 71-91%, '0' indications, 49-67%, 1 indication, 56-82%, 2 indications, 70-90%, and ≥3 indications, 97-99%. Conclusions: The cost-effectiveness of DES compared with BMS increased with increasing risk profile of patients from those who had 1, 2, to ≥3 indications for a DES. When compared with BMS, DES was least cost effective among patients with '0' indications for a DES. Based on these results, selective use of DES implantation is supported. These findings may be useful for evidence-based clinical decision-making.