Purpose:  Recent clinical trials have demonstrated the benefit embolic protection devices (EPDs) provide in reducing the risk of ischemic complications after percutaneous coronary intervention (PCI) of diseased saphenous vein grafts (SVGs).  However, little is known regarding the appropriate patient selection for these devices, including how the cost-effectiveness of embolic protection varies according to individual patient characteristics.  The objective of our study was to estimate the cost-effectiveness (CE) of using an EPD during SVG PCI, in a target population stratified by patient risk status.

Methods:  Control patients from 3 clinical trials (SAFER, X-TRACT, and FIRE) undergoing SVG PCI without embolic protection (n=738) were used to develop and validate a multivariate model to predict the occurrence of 30-day death or large myocardial infarction (MI) (CK-MB> 5x normal).  The 30-day costs associated with SVG PCI with and without complications were estimated using prospectively collected cost data from the SAFER trial population (n=801).  A Markov model was developed to determine the lifetime CE of using EPDs in patients defined as low, medium and high risk for 30-day death or large MI, from the perspective of the U.S. health care system.

Results:  EPDs become more cost-effective as a patient's risk for 30-day death or large MI increases.  In patients classified as low risk (5.3% event rate, 47% of patients), the use of an EPD was associated with an incremental CE ratio of $34,157 per quality-adjusted life year (QALY) gained.  This ratio dropped to $17,667 and $7,771 for patients defined as medium risk (10.8% event rate, 35% of patients) and high risk (26.7% event rate, 18% of patients), respectively.  The probability that EPDs are cost-effective at a threshold ratio of $50,000 per QALY gained is ³75% in each risk group (see Figure). 

Conclusion:  The use of EPDs is cost-effective across a wide range of risk subgroups and should be considered routine treatment for patients undergoing SVG PCI in the U.S.