Myocardial Perfusion Scans Projected Population Cancer Risks From Current Levels of Use in the United States

Background—Myocardial perfusion scans contribute up to 20% of the estimated annual collective radiation dose to the US population. We estimated potential future cancer risk from these scans by age at exposure and current frequency of use in the United States. Methods and Results—Usage patterns were determined from national survey data, and radionuclide dosage was based on current guidelines. Cancer risk projection models were generated on the basis of the National Research Council Biological Effects of Ionizing Radiation VII report, under the assumption that risk has a linear relationship with radiation exposure even at low doses. The mean projected number of radiation-related incident cancers and 95% uncertainty intervals were estimated with the use of Monte Carlo simulations. Estimated risks for a scan performed at age 50 years ranged from 2 cancers per 10 000 scans (95% uncertainty interval, 1 to 5) for a positron emission tomography ammonia-13 test to 25 cancers per 10 000 scans (95% uncertainty interval, 9 to 58) for a dual-isotope (thallium-201 plus technetium-99m) scan. Risks were 50% lower at age 70 years but were similar for men and women. The combination of cancer risk estimates and data on frequency of use suggests that the 9.1 million myocardial perfusion scans performed annually in the United States could result in 7400 (95% uncertainty interval, 3300 to 13 700) additional future cancers. Conclusions—The lifetime cancer risk from a single myocardial perfusion scan is small and should be balanced against likely benefit and appropriateness of the test. The estimates depend on a number of assumptions, including life expectancy. They apply directly to asymptomatic individuals with life expectancies similar to those of the general population. For individuals with a symptomatic clinical profile, on whom such scans are typically performed, the risks will be lower because of shorter life expectancy. (Circulation. 2010;122:2403-2410.)