Drug-eluting stents: a systematic review and economic evaluation

To assess the effectiveness and cost-effectiveness of the use of drug-eluting coronary artery stents in percutaneous coronary intervention (PCI) in patients with coronary artery disease.

Bibliographic databases, including MEDLINE, EMBASE and the Cochrane Library, were searched from December 2002 to August 2005. Hand-searching was also done.

A systematic literature review of effectiveness was conducted focusing primarily on randomised controlled trials (RCTs). Full economic evaluations that compared two or more options and considered both costs and consequences were eligible for inclusion in the economics review. A critique of manufacturer submissions to the National Institute for Health and Clinical Excellence and an economic evaluation in the form of cost-utility analysis were also carried out.

In the 17 RCTs of drug-eluting stents (DES) versus bare metal stents (BMS), no statistically significant differences in mortality or myocardial infarction (MI) were identified up to 3 years. Significant reductions in repeat revascularisations were determined for DES compared with BMS [for example, at 1 year: target lesion revascularisation (TLR) relative risk 0.24; 95% confidence interval (CI) 0.19 to 0.31; and target vessel revascularisation (TVR) relative risk 0.43; 95% CI 0.33 to 0.55]. This estimated benefit appears to be stable from 1 to 3 years. Binary restenosis and late luminal loss also favoured DES. In the eight RCTs of DES versus DES, no statistically significant differences in mortality or MI were detected between DES designs. In meta-analyses of TLR, TVR and composite event rate, marginal improvement in efficacy of Cypher trade mark over Taxus trade mark was observed. These results await confirmation beyond 1 year and differences in study design may have influenced reporting of outcomes. Ten full economic evaluations were included in the review and the balance of evidence indicated that DES are more cost-effective in higher risk patients. The review of submitted models confirmed the view that DES may be cost-effective only under very limited circumstances when realistic assumptions and data values were used. In the cost-utility analysis of DES versus BMS, the use of DES appears to reduce the rate of repeat revascularisations; benefit estimates used in the economic assessment are defined as 'broad' (i.e. cases involving any TLR/TVR irrespective of any other lesions/vessels undergoing revascularisation) and 'narrow' (i.e. cases involving TLR/TVR only). The incremental benefit to the patient is therefore described as the loss of quality-adjusted life-years (QALYs) avoided by not having to undergo a repeat revascularisation. Univariate sensitivity analysis and extreme values analysis indicate that the price premium, numbers of stents used in the index procedure and absolute risk reduction in repeat interventions most significantly influence the cost-effectiveness ratios. Sensitivity analyses also permit a range of values for efficacy and effectiveness to be considered for individual designs of DES. The cost-effectiveness results reveal that, all patients considered together, the calculated cost per QALY ratios are high (183,000-562,000 pounds) and outside the normal range of acceptability. Cost-effectiveness is only achieved for those non-elective patients who have undergone a previous coronary artery bypass graft and have small vessels. 'Real-world' data show that patient numbers in this latter group are very small (one in 3100 of all patients treated with PCI).

The conclusions of the assessment are that the use of DES would be best targeted at the subgroups of patients with the highest risks of requiring reintervention, and could be considered cost-effective in only a small percentage of such patents. This is similar to the conclusion of our previous assessment. Trials of DES compared with new generation BMS and with DES would be useful, as would further evaluation of newer BMS in combination with drug administration.