mgh logos Radiology Rounds
A Newsletter for Referring Physicians
Massachusetts General Hospital
Department of Radiology
Partners and Harvard logos

Volume 8 Issue 3 - March 2010
Print/Download this Article
Bookmark this Site
  Archived Issues of Radiology Rounds
MGH Department of Radiology Website
CT and Radiation Safety Issues

The estimations of risk of human exposure to ionizing radiation are largely derived from extrapolations of data from high radiation dose exposures experienced by Japanese atomic bomb survivors

Although substantially higher than that from conventional x-ray imaging, the radiation exposure associated with diagnostic CT imaging is considered to be low by the Health Physics Society

At MGH, a number of dose-reduction strategies are in place to minimize radiation exposure from CT imaging

Most CT scans are for older individuals and/or for those who already have cancer, for whom the risk is largely irrelevant; the risk for children is of more concern

The diagnostic benefits of CT imaging are well established and should be considered together with the potential risks

Estimating the Risks of Cancer
Minimizing the Risk by Minimizing the Radiation Dose
Further Information

  Figure 1. Chart showing numbers of excess deaths from solid tumors per 100 Japanese atomic bomb survivors (1950-1990) versus their dose. Error bars show 95% confidence limits. Note that four of the five data points are substantially higher than the radiation exposure from CT. Redrawn from Cohen, 2002.
  Figure 1. Chart showing numbers of excess deaths from solid tumors per 100 Japanese atomic bomb survivors (1950-1990) versus their dose. Error bars show 95% confidence limits. Note that four of the five data points are substantially higher than the radiation exposure from CT. Redrawn from Cohen, 2002.

Over the past 20 years, there has been dramatic growth in the utilization of CT, and around 72 million scans are now performed each year in the United States. Largely as a result of this growth, the radiation exposure associated with medical imaging now accounts for about 48% of the total exposure, up from 15% in 1987. Radiation doses from diagnostic CT can be several hundred-fold greater than that of conventional radiological procedures, such as a chest x-ray, and radiation from CT alone accounts for 24% of the total radiation exposure. There is no doubt that many patients have benefited from the rapid diagnoses made possible by CT and from its value for monitoring chronic disease. However, there is increasing concern regarding the risks of this exposure to radiation.

It is well established that radiation can be harmful and has both deterministic and stochastic effects. Deterministic effects, such as hair loss, skin burns, and cell death, are dose-dependent but do not occur below a threshold of 150-200 mSv. Since the typical estimated dose associated with proper use of CT is in the range of 2-10 mSv, deterministic effects are not normally a concern. Induction of cancer by radiation is a probabilistic (stochastic) effect, not a deterministic effect. That is, higher radiation doses are associated with a higher likelihood of carcinogenesis, but even low doses of radiation could potentially induce carcinogenesis and it is more difficult to assess a safe level of exposure.

Estimating the Risks of Cancer
The evidence that radiation induces cancer comes primarily from the longitudinal study of atomic bomb survivors in Japan (Figure 1). This study has collected data from approximately 100,000 individuals who experienced a one-time exposure from atomic bomb radiation and has been ongoing for over 60 years. In the Biological Effects of Ionizing Radiation (BEIR) VII report and that of the International Commission on Radiological Protection (IRCP), the risk of harm is estimated by using linear extrapolation of data from high radiation doses, the linear no-threshold (LNT) model. This model makes the assumption that there is no threshold below which there are no risks from radiation and assumes that the risk of radiation-induced cancer is about 0.05% per 10 mSv. However, it should be noted that there are some who dispute this conservative model and cite evidence of lower cancer rates in populations that have been exposed to low doses of radiation (hormetic effects). In addition, there is no direct evidence to date that CT is linked to cancer because of the long latency period of 30-40 years for solid tumors to develop and because of the difficulty of detecting an elevated risk at the relatively low levels of radiation exposure associated with CT when even without medical radiation exposure, one in three to one in four individuals in the U.S. will develop cancer in their lifetime. Nevertheless, it is wiser to be cautious and to assume that there is a risk, albeit small, from the radiation associated with CT scans. The small incremental risk of a potential radiation induced cancer should be weighed against information that CT can provide in a wide variety of appropriate clinical indication, a task aided by the MGH Radiology Order Entry Decision Support system.

Using the LNT model, it is estimated that the population risk of developing cancer is 4-6%/1000 mSv. The BEIR VII report estimated that the individual risk is approximately one in 1,000 for an adult who receives an abdominal CT and one in 550 for children and women. However, it should be noted that most solid cancers have a latency period of 30 years or more and that the majority of CT scans are for older individuals and/or for those who already have cancer, for whom the risk is largely irrelevant. Overall, the risks are low for the individual compared to the lifetime risk of developing cancer from all causes. However, the effect on overall public health could be significant; one recent projection concluded that as many as 29,000 future cancers could be related to CT scans performed in 2007. These risks must be balanced against the known diagnostic benefits of CT (Table 1). Patients need to be aware that one can point to individuals who have been harmed by not having CT when applied by accepted standards, but not vice versa.

Table 1. Typical Radiation Dose from Medical Imaging
Year Lateral chest x-ray Chest CT Abdomen & Pelvis CT Head CT
Historical benchmark 0.15 mSv 10 mSv 20 mSv 2 mSv
2010 (at MGH) 0.15 mSv 4 mSv 7 mSv 1.5 mSv

Minimizing the Risk by Minimizing the Radiation Dose
For some years, radiologists have believed that the best strategy is to image using the ALARA (as low as reasonably achievable) principle. There are a number of strategies for minimizing the radiation dose associated with CT. Some of these strategies, including tube-current modulation, cardiac gating, applying specific protocols for pediatric patients, and iterative reconstruction techniques have been described in previous issues of Radiology Rounds, Minimizing CT Radiation Dose and Iterative Reconstruction Methods.

Some tissues, particularly those that are growing, are more sensitive to radiation effects. Because of the greater potential of harm to children, Mass General Imaging has some specific strategies designed for this population. The radiation dose for a CT study is modified according to the patient’s size, the clinical indication, the body region, and whether there has been a prior CT. At the time of order entry, the information system knows these factors and can recommend the most appropriate scanning protocol. In addition, each request for pediatric CT is pre-screened by a radiologist and each patient is weighed prior to imaging to further optimize dose. The aim is to use the lowest dose to obtain a diagnostic quality image but not necessarily the “prettiest” image. Similarly for adult patients, body weight is a factor in the selection of radiation dose, with different sets of protocols for patients <135 lbs, 135-200 lbs, and >200 lbs.

Another strategy to reduce radiation exposure is to avoid unnecessary CT scans or to use alternate imaging modalities. The Radiology Order Entry (ROE) system at Mass General Imaging has recently been upgraded and now automatically detects duplicate orders, thus avoiding some unnecessary scans. In the case of pregnant patients, there is concern about the growing fetus. Therefore, only emergent CT is acceptable, in which case shielding is put in place and special protocols used. Alternatively, MRI or ultrasound will be used if those methods are likely to yield a diagnosis.

In the case of patients over 50 years or for those who already have cancer, the risks associated with CT are negligible because of the long latency period of cancer development which could span to 30-40 years for radiation-induced solid organ cancers. For all other patients, it must be recognized that the risk to the individual is low compared to many other risks in life and that the benefits of prompt diagnosis are real. However, ordering physicians must remember that adults are at lower risk of developing radiation induced cancer compared to children.

CT can be performed at Mass General Imaging in Waltham, Mass General Imaging Chelsea, Mass General/North Shore Center for Outpatient Care, or the main MGH campus and can be ordered online via the Radiology Order Entry (http://mghroe) or by calling 617-724-XRAY(9729).

Further Information
For further questions about CT radiation dose, contact Sanjay Saini, MD , or Mannudeep K. Kalra, MD Mass General Imaging, at 617-726-3043.

We would like to thank Mannudeep K. Kalra, MD, Sanjay Saini. MD, Mass General Imaging, and Timothy G. Ferris, MD, Internal Medicine, Massachusetts General Hospital, for their advice and assistance in the preparation of this article.

This article provided useful information about the appropriate use of imaging studies:

Note: clicking one of these options will close this window.


Amis ES, Butler PF, et al. (2007). American College of Radiology White Paper on Radiation Dose in Medicine. J Am Coll Radiol 4: 272-284.

Berrington de Gonzalez A, Mahesh M, et al. (2009). Projected cancer risks from computed tomographic scans performed in the United States in 2007. Arch Intern Med 169: 2071-2077.

Cohen BL (2002). Cancer risk from low-level radiation. AJR Am J Roentgenol 179: 1137-1143.

Little MP, Wakeford R, et al. (2009). Risks associated with low doses and low dose rates of ionizing radiation: why linearity may be (almost) the best we can do. Radiology 251: 6-12.

Miller JC (2008) Minimizing CT Radiation Dose. Radiology Rounds 6(2).

Miller JC (2009) Iterative Image Reconstruction Method (ASIR): Lowering CT Radiation Dose and Improving Image Quality. Radiology Rounds 7(10).