CT Scans and Radiation Risk: Caveat Interlocutor

Richard C. Semelka, MD


August 18, 2006

This edition of Semelka's Spin is the latest reprisal in the recent email give-and-take between yours truly and an extremely knowledgeable radiologist regarding the radiation risks associated with CT scanning.

This clinician takes issue with my interpretation of the BEIR VII report and the position of the authors of the report and other sources alleging that imaging x-rays cause cancer.

Peter Ronai, MD, PhD: In a recent posting on Medscape Radiology, "Imaging X-rays Cause Cancer: A Call to Action for Caregivers and Patients" (, you state that "... there is a 1 in 1000 chance of developing cancer from a single pass of a body CT." What is your evidence for this?

Richard Semelka, MD: Review the BEIR VII report, which states that there is a 1 in 1000 chance of developing cancer from exposure to 10mSv radiation.

Dr. Ronai: BEIR VII states: "At doses of 100 mSv or less, statistical limitations make it difficult to evaluate cancer risk in humans." As I understand this, BEIR conclusions are based on worst-case scenarios. BEIR VII does not explain why Denver Metro area cancer rates tend to be lower than national cancer rates (, despite the fact that the background radiation dose in Denver is more than triple the national average background radiation dose. Nor does it explain why average annual age-adjusted cancer incidence rates per 100,000 population are significantly lower in Colorado for all races than for the USA as a whole [Cancer in Colorado: 1996-2001, Colorado Central Cancer Registry], despite the fact that the majority of the Colorado population lives at a much higher altitude than does most of the US population.

In my opinion, the statement that "there is a 1 in 1000 chance of developing cancer from a single pass of a body CT scan" is an estimate (read "guess"), and the supporting data are extremely tenuous.

Dr. Semelka: This is a good point you raise. Regarding Colorado, there are of course a number of environmental factors that may contribute to cancer. If you read through everything that I have written for Medscape Radiology and elsewhere on this topic, I describe it as estimates and not concrete fact. However, I am sure you agree that it is well established that x-rays cause cancer, a fact that is beyond dispute; it seems that your concern is that the minimum dose is not known, and designating too small a dose as a minimum may be erroneous.

My primary point, if the minimum dose to result in cancer is not known beyond a reasonable doubt, is it not reasonable to err on the side of caution, and in fact present these points to patients and let them decide? Especially in the circumstance in which they are scheduled for CT scans for a medical indication, such as the assessment of liver lesions, when a better test, MRI, is available, and appears to be a much safer study.

Furthermore, an issue that has largely gone unaddressed in the x-ray debate is that highly fractionated exposure to any pathogen, whether it be infectious (eg, bacterial), substance-related (eg, inhalational exposure), or radiation-related -- is much less deleterious than a single large exposure.

In fact, the body probably can strengthen its defenses against an entity when it is presented in a highly fractionated form -- which may be what is happening in Colorado. You did not provide me with the annual radiation exposure in Colorado vs the rest of the United States -- I expect, without looking it up, that it may be 7 mSv vs 5 mSv, which, in a highly fractionated exposure (ie, daily), is negligible.

In the same way, I think it is inaccurate, what I myself describe for chest CT, that the 10mSv exposure is 1000 times greater than a chest x-ray of 0.1mSv -- in fact, it is much worse than 1000 times a chest x-ray because it is a sudden bolus of radiation. Radiation, as you know, causes double-stranded DNA breaks, and it is reasonable to conjecture that a fewer number of these breaks can be readily repaired or disposed of, whereas a larger bolus exposure may overwhelm the repair system.

This is analogous to multiple exposures to small quantities of streptococcus or tuberculosis, or inhalational smoke or silicosis, which can be dealt with by the immune system, whereas a sudden large exposure likely will overwhelm the system and result in serious harm and/or death. A simple, yet perhaps accurate comparison: If 80 degrees F is a comfortable temperature for us to live in, it is easier to compensate for a gradual 200-degree increase in temperature if it is delivered as 2 degrees hotter each day for 100 days, than a sudden blast of 200 degrees, which would result in death.

Dr. Ronai: I have no argument with your position that we should act judiciously with respect to the use of ionizing radiation and that we should weigh the potential risks and benefits. I have advocated this position for 45 years. I also agree with the principle that diagnostic imaging procedures that use nonionizing radiation should be employed in preference to those that use ionizing radiation, where the potential diagnostic yield of the former is equivalent or superior to that of the latter, particularly in the case of children and young adults.

My only objection is to the statements in the BEIR reports that give the impression that precise cancer equivalents are known for low radiation doses. I still maintain that these statements are best guesses based on worst-case scenarios.

Informed consent is a bedrock principle of medicine; however, there is informed consent and "informed consent." Patients cannot make a rational decision when faced with alarmist statements such as: "There is a 1 in 1000 chance of developing cancer from a single pass of a body CT."

I would prefer that patients be counseled in a much more nuanced way, such as: "At high-dose levels, x-rays are known to cause cancer, but at the substantially lower-dose levels associated with diagnostic x-ray procedures, the association between x-rays and cancer is much less clear. It has been estimated that there may be up to a 1 in 1000 chance of developing a cancer as a result of a body CT scan, but that is a best guess only, based on worst-case scenarios. At the relatively low radiation levels associated with most diagnostic x-ray procedures, proof of such an association would require a research study involving millions of subjects. Such a study has not been performed because of the huge logistic problems it would entail. In the meantime, it is prudent to use diagnostic x-ray procedures only when we anticipate a real diagnostic benefit that outweighs the potential theoretical risk, and when no alternative diagnostic procedure can provide equivalent information," or words to that effect, chosen according to the level of the patient's ability to absorb such information.

It was my practice (I am now retired) to supplement a verbal explanation with written material pertaining to informed consent or patient instructions, as experience has shown that even sophisticated patients have very limited ability to absorb complex concepts under the stress of a medical consultation. Having a clear written statement for patients to review at their leisure is a much more effective communication technique.

Background radiation doses for the United States as a whole are 3 mSv/yr for natural sources and 0.6 mSv/yr for artificial sources for a total of 3.6 mSv/yr ( Background radiation doses for Denver are 11.8 mSv/yr for natural sources and 0.6 msSv/yr for artificial sources for a total of 12.4 mSv/yr (

You are correct that the biological effects of acute radiation exposure are greater than those of chronic exposure. However, the usual clinical situation in which CT scanning might introduce radiation-related concerns is neither classically acute nor chronic, but rather "repeated low level acute" (as in several CT scans over a 12-month or longer period, for example). Therefore, one could postulate that this would lead to a risk profile that is intermediate between classically acute and chronic exposures.

I have no objection to your submitting our correspondence to Medscape, provided my position is accurately represented. I am concerned by your statement "to show points on the opposite side of the debate." I am not really on the opposite side of the debate. I agree fully with your basic position. I just disagree with the use of statements of certainty when no certainty exists.

Dr. Semelka: I agree with everything that you have written. I think how you have worded your information to patients is both nonalarmist and accurate, and may be the most effective form of dialogue to have with patients.

To editorialize on this dialogue, it became progressively clear in my correspondence with Dr. Ronai that he was highly knowledgeable about radiation risk. His discussion of background radiation introduces another point related to background radiation, which is radiation exposure from flying, which is estimated to be .05 mSv/hour. This means one would have to fly for 200 hours in order to get the radiation exposure of 1 body CT study (10 mSv/). I wanted to include this as a patient had emailed me that his physician informed him that 1 body CT was equivalent to 1 transcontinental flight, which is not true. One body CT equals 40 transcontinental flights -- that's a lot of flying. Of course, probably the number of hours is many more, because of the ameliorating effect of fractionation radiation of dose in flying, which would have much less impact than 1 bolus of 10 mSv radiation from CT.

Dr. Ronai: Few flyers accumulate that much flying time, but airline pilots certainly do. Several studies of cancer incidence in airline pilots have produced decidedly mixed results. The largest studies have come from the Scandinavian countries. (See, for example, Haldorsen T, Reitan JB, Tveten U. Cancer incidence among Norwegian airline pilots. Scand J Work Environ Health. 2000;26:106-111.) Haldorsen and colleagues followed 3701 male pilots over 70,560 person-years and came to the following conclusions: "For most cancer sites, the [cancer] incidence among pilots did not deviate from that of the general population and could not be related to block hours of flight time or dose. It seems more likely that the excess risks of malignant melanoma and skin cancer are explained by factors related to life-style [greater sun exposure during leisure hours] rather than by conditions at work."

In the largest, most powerful study, Pukkala and colleagues (Pukkala E, Aspholm R, Auvinen A, et al. Incidence of cancer among Nordic airline pilots over five decades: occupational cohort study. BMJ. 2002;325:567) came to the following conclusions:

Few areas outside the Nordic countries have a history of several decades of population based registration of cancer. The large joint cohort size and five independent country specific observations in this joint Nordic study reduce the possibility of chance findings, and carefully registered incidence data help to avoid artifacts possibly included in the mortality statistics. Because of the complete population registration systems in all Nordic countries and accurate computerized record linkage procedures, the standardized incidence ratio estimates in this study are not prone to bias attributable to incomplete follow up or failures in record linkages.

Because our cohort included most of the cockpit crew ever certified in the Nordic countries, this study can be considered to have maximum potential to study cancer incidence among pilots. We were able to assess cancer risk by level of exposure, take into account characteristics of cancer (for example, subtypes of leukaemia, tumour latency), and estimate independent effects of exposure, age, and time period of the diagnosis in a way that has not been meaningful in small national settings. Apart from skin cancers, male Nordic pilots seem to have a pattern of cancer typical of that of high social class men in the Nordic countries. Our study shows a need for detailed studies focusing on possible work-related factors involved in the evidently raised risk of skin cancer and the suggestive dose-response patterns in prostate cancer.

Dr. Semelka: It is very clear from this discussion that Dr. Ronai is well versed on radiation risk. If a national quorum is eventually established to discuss this subject, I would strongly advocate he be on the panel. My final response to his erudite description of flying associated with no clear increase in cancer is that, again, because of the fractionated nature of the dose delivery of cosmic radiation in flying, it is not strictly relatable to a single large bolus of radiation -- that is, 200 hours x .05 mSv (flying) is not equivalent to 10 mSv delivered in 5 minutes (CT).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.