HelpLine 1-800-298-2436

LCA Responds to NEJM Article on CT Scans

December 10, 2007

Background: A recent paper entitled Computed Tomography – An Increasing Source of Radiation Exposure by David Brenner PhD and Eric Hall PhD, was published by New England Journal of Medicine on November 29, 2007 and subsequently picked up by several news outlets.

The paper cited the increasing use of CT scans for a wide range of purposes and concluded: “Although the risks for any one person are not large, the increasing exposure to radiation in the population may (emphasis added) be a public health issue in the future.”  The paper does not prove that CT’s cause cancer and even notes that “most diagnostic CT scans are associated with very favorable ratios of benefit to risk.”

Sadly this was buried in the paper and missed by most of the media. Many of you have become justifiably concerned with references to the atomic bomb and cancer risk. Thus we are providing the following analysis so that patients and those at risk for lung cancer do not mistakenly avoid talking with their doctors about their own individual risks versus benefits of life-saving medical imaging care.

OVERVIEW: The article does sound legitimate warnings about the potential dangers of the overuse of CT scans, especially on children (for appendicitis, injuries, headaches etc.) where the cumulative lifetime radiation dose may be of concern.  LCA agrees with that.  No scan should be done without a good reason. There are risks and benefits to scans that will vary with each individual case and should be discussed by patients and their doctors.

PREMISE IN THE ARTICLE: Since the number of CT scans being given annually is escalating (26 million a year in the mid-nineties to 62 million a year now), so too is the risk of radiation induced cancer.

Comment by LCA: That is an assumption not a proven fact. And even if true, the increased risk they project would be the sum total for the population as a whole (the tree) and not for an individual (a leaf).  The risk and benefit of a scan will always vary with individuals.  There are some studies showing an increased cancer risk for higher doses of radiation (over 100 mSv). But, in another paper by the same author it is conceded that a sample size of 5 million people would be needed to test the premise that lower doses (10 mSv) would lead to a statistically significant increase in cancer risk.

PREMISE: CT scans give higher radiation doses than x-rays.

Comment: True. Dose is a factor of the amount of radiation, time of exposure and density of the area. A flat (two dimensional) x-ray of the chest gives a dose of about 0.1 millisieverts (mSv).  A chest CT scan is an x-ray given in a spiral around the chest that yields a three dimensional image. Initially a chest CT was as high as 8 to 10 mSv but thanks to advances in both CT technology and in the development of least possible dose screening protocols, a typical chest CT is now about 1.5 mSv. Indeed, it is called a low dose CT scan.                    

That’s 15 times the dose of a chest x-ray. But, consider this. Even setting aside the well documented and universally accepted superiority of CT scanners to diagnose lung cancer at its earliest stages, let’s put that number into perspective. The following are approximate mean, individual dose estimates from a 2003 paper on the same topic by the same author (these estimates were not included in the new paper):

Average normal background radiation dose:               3 mSv per year
Single screening mammogram (breast dose)            3 mSv
Pediatric CT scan (stomach dose)                                 25 mSv
Radiation worker exposure                                              20 mSv/year
Exposure on space station                                              170 mSv/ year
A bomb survivors (mean dose)                                       200 mSv

PREMISE: The calculations in the paper are based on an average CT dose to a specific organ of 15 mSv.

Comment: That amount is 10 times the typical chest CT (1.5 mSv).  Dropping their estimates by a factor of 10 would render them insignificant.

PREMISE:  “There was a significant increase in the overall cancer risk in the subgroup (emphasis added) of atomic-bomb survivors who received low doses of radiation, ranging from 5 to 150 mSv; the mean dose in this subgroup was about 40 mSv which approximates the relevant organ dose from a typical CT study involving two or three scans in an adult.” (No figures on the number of observed cancer cases were given.)

Comments:  One, atomic-bomb survivors were exposed, in addition to x-rays to the whole body, to particulate radiations, neutrons and other radioactive material that people receiving CT scans are not exposed to.

Second, as noted above, the mean exposure (half received more and half less) for all the atomic-bomb survivors being studied was 200 mSv. In a previous paper the author demonstrated that the increased risk for the subgroup of people who received 5 to 100 mSv (with a mean dose of 29 mSv) was statistically insignificant. In this paper he turns insignificant into slightly significant by expanding the top range of the subgroup to 150 mSv (with a mean dose of 40 mSv). He states that the whole subgroup of 5 to 150 mSv being is 26,000 and we can assume the 5 to 100 mSv subgroup is even smaller - in either event far below the millions of people the author concedes in a previous paper would need to be studied for a valid assessment of increased risk at lower doses.

Third, Since the typical chest CT is about 1.5 mSv to reach threshold of 40mSv for possible increased risk would require at a minimum 26 scans.

Fourth, the author also states that in any event, the possibility of increased risk decreases with age and drops significantly after age 55.  Most lung cancer patients are older. Those at high risk for lung cancer over the age of 50 are urged to talk with doctors about a CT scan.  Those with suspicious nodules requiring 2 to 3 follow-up scans would receive a combined dose of less than 5 mSv.

PREMISE: If all the above contested factors and risk estimates were accepted as correct and valid, then 1.5 to 2% of all cancers may be attributable to CT scans.

Comments: That is, about 1 in 60 of the population as a whole MAY be at risk for radiation induced cancer provided that they receive at least 40 mSv of radiation.   To put that in context, according to SEER/NCI figures men in the US currently have a  1 in 2 risk of developing any cancer in their lifetimes and women 1 in 3.  The lifetime risk of developing lung cancer is 1 in 13 for men and 1 in 17 for women.

PREMISE: The paper makes three recommendations: reduce the dose per scan (the authors concede that this has already been done with new scanners); substitute MRI’s, ultrasound and other procedures whenever practical; reduce the number of excess scans being done, especially on children.

Comment: LCA agrees.

Advisories Released by Other Organizations:

1. American Society of Clinical Oncologists (ASCO)

Expert Perspective from ASCO on the Link Between Cancer Risk and the Increased Frequency of CT Scans
A review article, Computed Tomography -- An Increasing Source of Radiation Exposure, published in the November 29, 2007 issue of The New England Journal of Medicine asserts that the radiation from computed tomography (CT) may cause up to 2% of all cancers in the United States.


A CT scan creates a three-dimensional picture of the inside of the body using a series of x-ray pictures that are taken from many different angles. A computer compiles these images into a detailed, cross-sectional view. Because CT scans have proven so useful and beneficial in the practice of medicine, the use of CT has steadily increased. In the article, the authors cite data estimating that 62 million CT scans a year are done in the United States.

The fact that radiation exposure increases cancer risk has been well-studied, especially in atomic bomb survivors. It is well known that diagnostic x-rays and CT expose people to low levels of radiation. The concern raised in this study is that the amount of radiation delivered through CT is many times higher than any other radiology test.

Using published data, the authors estimated how much radiation is delivered by a CT examination. By applying data from the atomic bomb exposure studies, the authors extrapolated the number of cancer cases that might result from CT exposure. With the increasing use of CT, 1.5% to 2% of all cancers theoretically may be caused by CT, the authors concluded.

Additional perspective

"This is a compelling article that raises interesting and important issues," said Allen S. Lichter, MD, Executive Vice President and CEO of ASCO. "Although some are questioning the methodology of this article, it is important to reduce cancer risks where possible, and the recommendations in the article are reasonable."

The recommendations mentioned in the article include lowering the CT-related dose for each patient, replacing CT use, when practical, with other options, such as ultrasonography (ultrasound) and magnetic resonance imaging (MRI, a test that uses electromagnetic waves to create pictures of the inside of the body), and decreasing the number of CT studies that are prescribed.

Bottom line

"This review reminds us that doctors and patients must always consider the possible risks and benefits--both immediate and long term--for any medical intervention. For patients with active cancer, CT scanning is usually beneficial, but we should always consider how it will enhance care and whether other tests that involve less radiation exposure can be substituted," said Nancy Davidson, MD, ASCO President and Director of the Breast Cancer Program at the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins University in Baltimore, Maryland.

"The most important message is that when a CT scan is needed for diagnosis, especially for the management of cancer, the benefits always exceed the risks. But, other imaging methods should be used when they can provide the same benefit with less radiation exposure," added Gabriel Hortobagyi, MD, FACP, ASCO’s Immediate Past President and Chair of the Department of Breast Medical Oncology at the University of Texas M. D. Anderson Cancer Center in Houston.

Dr. Lichter agreed, "CT can save lives and in the cases where it is important for patients' medical care, the benefit and information gained far outweighs the risk. CT is certainly getting better as technology improves and lower doses are being used. However, CT should only be used when medically necessary."

What This Means for Patients:

The American College of Radiology (ACR) has information for both physicians and their patients at and recommends that patients keep a record of their x-ray and CT scan history and talk with their doctor about the rationale for the CT scan if one is recommended. The ACR also provides accreditation of facilities with x-ray technology and defines criteria for providing the most appropriate imaging scan. Additionally, the Radiological Society of North America (RSNA) has relevant information about the use of CT and other imaging tests at

ASCO Member Alert: Increased Frequency of CT Scans May Raise Cancer Risk

2. American College of Radiology

ACR Responds to NEJM Article on Radiation Risk Associated With CT Scans

The American College of Radiology (ACR) is concerned that certain conclusions and comparisons made in the study, “Computed Tomography — An Increasing Source of Radiation Exposure,” published in the Nov. 29 issue of the New England Journal of Medicine, may be inappropriate and cause patients to mistakenly avoid getting life-saving medical imaging care.

The study claims that up to 2 percent of all cancers in the United States may be caused by radiation received from CT scans. Yet, the study authors admit that there are currently no published studies directly linking CT scans (even multiple CT scans) to cancer.

The study also equates radiation exposure and effects experienced by many atomic bomb survivors in Japan to present day patients who receive computed tomography (CT) scans. Most CT exams are performed in a controlled setting. They result in limited radiation exposure to a small portion of the body. Atomic bomb survivors experienced instantaneous radiation exposure to the whole body. Also, CT exams expose patients solely to X-rays. Atomic blast survivors were exposed to X-rays, particulate radiations, neutrons, and other radioactive materials. The known biological effects are very different for these two scenarios.

“Patients need accurate information on which to base their health care decisions. They may be terribly confused and unduly distressed by some of the statements in this study,” said Arl Van Moore Jr., M.D., FACR, chair of the ACR Board of Chancellors.

There is little doubt in the medical community that CT scans help save lives. Advancing technology has increasingly allowed imaging exams to replace more invasive techniques, but has also resulted in increased radiation exposure for Americans.

The College urges patients and providers to visit the “Radiology Safety” section of the ACR Web site as well as the “Radiation Safety” section of, the patient information site co-managed by the ACR and the Radiological Society of North America (RSNA), for more information regarding radiation exposure from medical imaging exams.

Patients should also keep a record of their X-ray history and before undergoing a scan, should ask their physician:

  • Why do I need this exam?
  • How will having this exam improve my health care?
  • Are there alternatives that do not use radiation which are equally as good (e.g. MRI, ultrasound, etc.)?
  • Is this facility ACR accredited? (This ensures that physician and staff meet   education and training standards and that equipment is surveyed regularly by medical physicist to ensure that it is functioning properly.)

“The College has long held that no medical test, particularly those utilizing ionizing radiation, should be performed unless the medical benefits clearly outweigh any risk associated with the exam. For example, the ACR has long opposed full body CT scans for asymptomatic patients, one of the exams that the NEJM article authors put forth as a driver of future CT growth,” said ACR Chair Moore. We also support the ‘as low as reasonably achievable’ (ALARA) concept which urges providers to use the minimum level of radiation needed in such exams to achieve the necessary results.

ACR Responds to NEJMA Article on Radiation Risk Associated with CT Scans.