Measuring the lifetime costs and benefits of medical technologies is essential in evaluating technological change and determining the productivity of medical care. Using data on Medicare beneficiaries with a heart attack in the late 1980s and 17 years of follow up data, I evaluate the long-term costs and benefits of revascularization after a heart attack. I account for non-random selection into treatment with instrumental variables; following McClellan, McNeil, and Newhouse, the instrument is the differential distance to a hospital capable of providing revascularization. The results show that revascularization is associated with over 1 year of additional life expectancy, at a cost of about $40,000. Revascularization, or other treatments correlated with it, appears to be highly cost-effective.