Radiation Exposure from Diagnostic Tests
Cancer patients often get worried that the multitude of diagnostic
scans may put them at higher risk of getting a second cancer. These
worries are exacerbated by news accounts that compare the radiation
levels in a CT scan to those experienced by survivors of Hiroshima.
Generally, the potential risk of a second cancer from X-rays and CT scans is much less than the actual harm of failing to adequately stage the known cancer or detect a relapse.
Note that only X-rays and CT scans involve a form of ionizing radiation. Ultrasounds and MRIs do not involve radiation. Ultrasounds use low frequency noise to measure tissue density. MRIs use magnetic fields and radio waves to cause protons in hydrogen atoms (within water molecules) to resonate.
There is not much good information about radiation exposure levels in the literature. Most of what is reported is estimated using questionable techniques. None are based on direct measurement. The figures we report below are among the levels that seem to be more reliable, but are still questionable.
Natural exposure to background radiation in the US ranges from 240 to 480 millirems per year, depending on altitude and radon exposure. The average annual exposure is about 350 millirems.
These figures do not count exposure from watching television or sitting in front of a computer CRT, which has a Federal limit of half a millirem per hour.
For occupational exposure the safe limit is 5,000 millirems per year.
A coast-to-coast airplane flight involves exposure of about 30 millirems.
During a chest x-ray, the patient is exposed to 20 to 40 millirems of radiation. This is similar to the amount of radiation exposure that occurs in a coast-to-coast airplane flight.
A chest CT using an electron beam CT scanner involves 125 to 160 millirems. (Abdomen + Pelvis adds 160 to 200 millirems.)
A chest CT using a helical CT scanner involves 500 to 700 millirems. (Abdomen + Pelvis adds 800 to 1,100 millirems.)
Factors affecting the dosage include whether the X-ray source is on continuously or just when images are being taken (a conventional CT does the former while an electron beam CT does the latter), the number of detector rings, and the length of the procedure. Diagnostic equipment that involves digital imaging instead of exposure of film can reduce the amount of radiation exposure by a factor of 10.
There is a common myth that a CT scan is 100 times the exposure of an x-ray. This myth is based on the faulty assumption that one slice of a CT scan is the equivalent of a chest x-ray, and counts the number of slices. The actual density of radiation exposure is much lower than that.
Modern high resolution CT scanners are also more efficient, limiting the total exposure of the patient. They use more sensors with noise-filtering software and dynamic current control to achieve higher resolution at a lower dose. New x-ray tubes and shaped filters also limit scatter radiation.
The amount of radiation used in radiation therapy is approximately 5 orders of magnitude higher than that in an x-ray.
A handful of papers have compared radiation exposure from diagnostic imaging with the levels experienced by survivors of Hiroshima. These papers include:
These analyses involve several flawed assumptions:
The lowest dose of Japanese survivors of the atomic bomb was about 2,000 millirems. Although there was a slight increase in the relative risk of cancer among the survivors, this includes survivors with much greater exposures. So comparing their exposure with CT scans is a bit like comparing apples and oranges.
The bottom line is that although unnecessary radiation exposure should be avoided, you should not fret if your oncologist feels you should have a CT scan. The radiation exposure from a chest CT is only 3-4 times that of a chest X-ray, and the added risk is justified by the benefits in terms of improved outcomes. The overall increase in lifetime cancer risk from a full-body CT scan is less than a tenth of one percent.
Copyright © 2005-2009 by Mark Kantrowitz. All rights reserved.
Suggestions and corrections are welcome and should be sent to .