Abstract Archives of the RSNA, 2010
John D. Keen MD, MBA, Presenter: Nothing to Disclose
James E. Keen PhD, DVM, Abstract Co-Author: Nothing to Disclose
Between the ages of 30 and 50, about 1000 single-phase CT scans of the abdomen and pelvis will induce one future cancer over a lifetime. Surveillance Epidemiology and End Results (SEER) data show half of all cancers are lethal, so the absolute death risk is 0.5/1000, the same as the absolute benefit from 10 years of routine screening mammography. Given widespread innumeracy, we wanted to provide another perspective to promote insight into the radiation harm from the typical CT scan.
We obtained the projected number of future cancers related to overall CT scan use in the United States in 2007 by age at exposure from a recent analysis. Since the cancer types were not broken down by race or sex, we calculated combined life expectancy estimates by using the gamma-mixed exponential (GAME) method, which required the United States Life Table for 2005 and SEER 10 year survival statistics for each cancer diagnosed in 1996. We assumed CT exposure at age 30 and age 50, with worst-case cancer induction at 5 years as well as at 20 years for age 50. We calculated years lost in life expectancy along with a utility loss assuming a chronic cancer state of 88% of normal health.
At age 50, the normal life expectancy is 30 years. The top five lethal cancers for life expectancy are pancreas 1.2, liver 1.8, lung 2.9, esophagus 3.5, and stomach 4.9 years. With equal cancer induction, the years lost are 13.9 and utility loss is 1.4 years. With CT weighted cancer induction, the total life loss is 16.7 years. Lung cancer contributes 30%, followed by stomach at 13%, leukemia and colon at 10%, liver at 7%, oral at 6%, bladder and brain at 5%, pancreas at 4%, and breast at 3%. For cancer induction at 5 years, the loss averages 3.6 days for every CT scan. At 20 years induction, the loss averages 8.9 years per cancer and 1.9 days per scan. At age 30 and life expectancy of 49 years, the loss averages 30 years per cancer and 7.4 days per scan. We did not discount the results.
For a CT scan at age 50, the percentage total life loss from a CT induced cancer at worse is 16.7/30.5 years or 54%. Every CT patient between the ages of 30 and 50 loses on average between 2 days and 1 week of life expectancy.
Since the average life gain per mammogram for 40-year-old women routinely screened for 10 years is about 1 day (undiscounted), should radiologists inform their patients that CT scans take lives?
Keen, J,
Keen, J,
Estimate of Life Expectancy and Utility Loss from Computed Tomographic (CT) Scan Radiation: A Different Perspective to Support Consumer-oriented Medical Decision Making. Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL.
http://archive.rsna.org/2010/9001817.html