SPECIAL ARTICLE
Answers to Common Questions About the Use
and Safety of CT Scans
Cynthia H. McCollough, PhD; Jerrold T. Bushberg, PhD; Joel G. Fletcher, MD;
and Laurence J. Eckel, MD
Abstract
Articles in the scientific literature and lay press over the past several years have implied that computed
tomography (CT) may cause cancer and that physicians and patients must exercise caution in its use.
Although there is broad agreement on the latter pointdunnecessary medical tests of any type should
always be avoideddthere is considerable controversy surrounding the question of whether, or to what
extent, CT scans can lead to future cancers. Although the doses used in CT are higher than those used in
conventional radiographic examinations, they are still 10 to 100 times lower than the dose levels that have
been reported to increase the risk of cancer. Despite the fact that at the low doses associated with a CT scan
the risk either is too low to be convincingly demonstrated or does not exist, the magnitude of the concern
among patients and some medical professionals that CT scans increase cancer risk remains unreasonably
high. In this article, common questions about CT scanning and radiation are answered to provide phy-
sicians with accurate information on which to base their medical decisions and respond to patient
questions.
ª 2015 Mayo Foundation for Medical Education and Research n Mayo Clin Proc. 2015;90(10):1380-1392
O
ngoing developments in computed years is that some patients forego critically
tomographic (CT) imaging have led needed CT examinations because of the belief
From the Department of to an ever-increasing number of clini- that these examinations are more harmful than
Radiology, Mayo Clinic,
Rochester, MN (C.H.M., J.G.F.,
cal applications, many of which have supplanted beneficial16; this problem will likely be exacer-
L.J.E.); and Department of less accurate or more invasive diagnostic tests. bated by the Times’ op-ed piece and by overinter-
Radiology, UC Davis School For example, CT has the highest sensitivity pretation of recent epidemiological studies.17,18
of Medicine, Sacramento, CA
(J.T.B.).
(95%) and specificity (98%) for urinary stone Similarly, some physicians refrain from ordering
detection than does any imaging technique, medically appropriate CT examinations
including radiography and ultrasound,1-8 and because of well-intentioned, but misinformed,
CT angiography has almost replaced invasive concerns.16 In this article, we address common
angiography as the initial test of choice. Because questions about the use and safety of CT to
of their clinical value, the number of CT scans ensure that physicians are equipped with credible
performed annually in the United States has information on which to base their decisions
increased substantially; an estimated 81 million when weighing the risks and benefits of ordering
CT scans were performed in the United States CT scans.
in 2014.9 Although there is a perception among
some physicians and patients that the dose of HOW IS RADIATION DOSE IN CT
ionizing radiation from medical imaging exami- QUANTIFIED?
nations, particularly CT, poses a substantial can- Several radiation dose metrics are currently used
cer risk to patients, this perception is not in CT dosimetry, each of which is used for
consistent with data from high-quality studies, different purposes. The volume CT dose index
nor with current consensus opinions of radiation (CTDIvol, reported in units of milligray) is one
protection organizations.10-14 In a recent op-ed commonly used metric that is displayed on the
article in the New York Times,15 2 physicians CT scanner console or in patient dose reports.19
expressed their opinion that CT examinations The volume CT dose index is useful for
performed in the United States and elsewhere describing the radiation output from a CT scan-
are “killing people.” A growing problem in recent ner and optimizing CT protocol parameters.
1380 Mayo Clin Proc. n October 2015;90(10):1380-1392 n http://dx.doi.org/10.1016/j.mayocp.2015.07.011
www.mayoclinicproceedings.org n ª 2015 Mayo Foundation for Medical Education and Research
, USE AND SAFETY OF CT SCANS
However, CTDIvol does not represent the pa-
TABLE. Typical Effective Dose Values Associated With Various Medical Imaging
tient’s absorbed dose.20 Estimates of patient
Examinations, Background Sources of Ionizing Radiation, and Regulatory
dose must take the patient’s body habitus
Limitsa,b
into account. The American Association of
Source of radiation exposure Examination Effective dose (mSv)c
Physicists in Medicine developed a method
to calculate size-specific dose estimates Radiography and fluoroscopy Hand radiograph <0.01
Dental bitewing <0.01
(SSDEs) using the reported CTDIvol values
radiograph
and a measure of patient size.21 The size-
Chest radiograph 0.02
specific dose estimate calculates the mean Mammogram 0.4
absorbed dose at the center of the scan range. Lumbar spine radiograph 1.5
For organs fully contained in the scan range, Barium enema 8
the SSDE provides reasonable approxima- Fluoroscopic coronary 7
tions of organ doses. For reference the brain angiogram
dose from a head CT scan is approximately Computed tomography Head CT 2
50 to 60 mGy and the colon dose from an Chest CT 7
Abdomen CT 8
abdomen/pelvis CT scan is approximately
Pelvis CT 6
15 to 20 mGy. Coronary artery 3
Because much of the data on radiation risk calcification CT
involves exposure to the whole body (eg, from Coronary CT angiogram 16
studies of the survivors of the atomic bombings Radionuclide imaging Lung scan 2
in Hiroshima and Nagasaki, Japan), a mechanism Bone scan 4
for comparing partial body irradiations, such as Myocardial perfusion 14
in CT, with whole-body irradiations is desirable. imaging
Naturally occurring sources of
To accomplish this, a radiation protection quan-
ionizing radiation (eg, cosmic 1.3-9.6
tity known as effective dose is used.11 The effective
rays or radon gas) (US average¼3.0)
dose does not represent the individual biological Maximum allowable annual
risk to any particular patient, but rather is used to occupational dose to 50
compare the radiation risk from different types of radiation workers (US)
radiation sources and different imaging examina- a
CT ¼ computed tomography.
tions.22 Effective doses are reported in units of b
The values for effective dose presented here are typical values for examinations in adults. Vari-
millisieverts, and typical effective doses from ations from these values would be expected because of differences in body habitus (especially in
CT scans range from less than 1 to approximately young children and infants), details of the imaging protocols, and equipment used.
c
10 mSv. For reference, in the United States the Reliable estimates of risk cannot be attributed to effective doses below 100 mSv.
Data from The Essential Physics of Medical Imaging.23
average effective dose from naturally occurring
background radiation (eg, unavoidable environ-
mental exposures, such as radon gas and cosmic
rays) is approximately 3 mSv/y. (Table). Some procedures require multiple scans
When evaluating the “dose” from a CT scan, over a region; for example, examinations using
it is essential that one understand what type of iodinated contrast material to visualize tissue
dose is being discussed and then compare that vascularity may need to include scans during
dose to risk data appropriate for that dose both the arterial and venous phases. For exami-
metric. For example, the absorbed dose to the nations requiring multiple scans at different
brain from a head CT scan is approximately contrast enhancement phases, these individual
60 mGy, but the effective dose from the same scan doses can add up to 20 to 30 mSv. How-
CT scan is only approximately 1.5 mSv. ever, this total is still considered a low dose of
radiation, which is defined by the radiation pro-
HOW MUCH RADIATION DOES CT USE? tection and radiation biology communities as
A CT scan delivers an effective dose of anywhere dose levels below 100 mSv.
from less than 1 to around 10 mSv, depending In the United States, the annual effective
on the type of scan the patient receives. For dose from ubiquitous background radiation is
example, the exposure from a head CT scan is on average 3 mSv/y; the typical range is from
approximately 1 to 2 mSv whereas the exposure 1 to 10 mSv. In regions at higher elevation
from a body CT scan is approximately 10 mSv (which are exposed to more cosmic rays) or
Mayo Clin Proc. n October 2015;90(10):1380-1392 n http://dx.doi.org/10.1016/j.mayocp.2015.07.011 1381
www.mayoclinicproceedings.org
