gms | German Medical Science

Pressure-derived fractional flow reserve (FFR) can be used to assess the functional significance of intermediate coronary stenoses. Previous studies have found that use of FFR to identify patients with hemodynamically insignificant stenoses (FFR>0.75) is both clinically beneficial and cost-effective. However, the impact of drug-eluting stents (DES) on these clinical and economic outcomes is unknown.

The objective of our study was (1) to evaluate the cost-effectiveness of FFR testing with selective PCI versus universal PCI in patients with single vessel disease at angiography, mild-to-moderate angina, and no documented ischemia. Furthermore, we sought to compare results for the drug-eluting stent era with those for the bare metal stent (BMS) era.

We developed a decision model [Fig. 1] to predict the life expectancy, quality-adjusted life expectancy (QALE), direct lifetime costs, and incremental cost-effectiveness ratio (ICER) in costs per quality-adjusted life-year gained (QALY) for patients with mild-to-moderate angina, single vessel disease at angiography, but without documented ischemia who were scheduled for percutaneous coronary intervention (PCI). We compared two strategies: 1) Universal PCI (UNIV) without FFR testing and 2) FFR testing followed by PCI only for those with FFR<0.75 (FFR). The base-case analysis considered a 60-year-old man under the optimistic assumption (for UNIV) that relative mortality reduction with revascularization is independent of a lesion's functional significance. Effectiveness of FFR was based on the results of the DEFER trial [Ref. 1]. Long-term clinical outcomes of PCI and medical management including recurrence rates, disease progression, and quality of life were estimated from the published literature. Based on fixed effects meta-analysis, we estimated that DES reduce clinical restenosis rates by 79% compared with bare metal stents (BMS). We adopted a societal perspective and discounted costs and QALYs at 3% per year. We performed all analysis assuming (i) the stent was a drug-eluting stent and (ii) the stent was a bare metal stent.

For the case of BMS, UNIV increased costs by $2,800/patient and improved outcome by 12 quality-adjusted life days (QALD), yielding an incremental cost-effectiveness ratio (ICER) of $84,000 per quality-adjusted life year (QALY) gained. When considering the benefits of DES, the cost difference increased to $3,300 and the benefit of UNIV increased to 18 QALDs, with an ICER of $69,000/QALY. Results were similar for women and over a broad age range (55-75 years). If we assumed that PCI in functionally insignificant stenoses did not reduce long-term mortality, the ICER for UNIV vs. FFR was >$1 million/QALY. Further one- and two-way sensitivity analyses indicated robustness of these results. Figure 2 [Fig. 2] shows the combined results for DES and BMS.

Our study has several limitations. First, the effectiveness data were based on a single center trial with moderate sample size and limited time horizon (DEFER trial). Second, the natural history of 1-vessel coronary artery disease with and without functional stenosis not yet sufficiently investigated, Third, the mortality benefit of PCI in functional vs. non-functional stenoses is not known and was based on a conservative assumption favoring UNIV.

In conclusion, regardless of the type of stent (BMS or DES), measuring FFR to guide the decision to perform PCI leads to significant cost savings to the health-care system. Even under very optimistic assumptions regarding the mortality benefits and restenosis rates for DES, the universal stenting approach does not appear to be economically attractive by conventional standards. Long-term follow-up studies of patients with and without functional stenosis are needed to provide important information.