Image gently, or when the diagnostic is worse than the disease

From the time I was little I had an awareness of the potential dangers of medical X-rays. My mom was always careful to limit the number and frequency of dental x-rays she let us have, and luckily we rarely needed X-rays for anything else. Her concern was probably based on scientific awareness coupled with regret from playing with fluoroscopes in shoe-stores in the 1950s. I’m sure it also had something to do with the fact that a relative whom we saw a couple of times a year had a developmentally disabled daughter, most likely having something to do with working as an x-ray technician in the 1950s while pregnant.
So with the rise of imaging, especially high-radiation CT scanners, I’ve written about the possible dangers (see Saying no to CT). It’s shocking to me the number of CT scans that individuals receive, in some cases approaching or exceeding the radiation exposure from survivors of Nagasaki and Hiroshima.

I’m glad to see that the Alliance for Radiation Safety in Pediatric Imaging has announced an “image gently” campaign, designed to educate radiologists, technologists, medical physicists and parents to minimize radiation dose in children. They have an excellent website, which I recommend. See for example, the section for parents, which advises on how to be an advocate for your child. Even if you are shy about taking an active role in your own care, I urge you to do so when it comes to your kids.
The site lists five steps that radiologists should take to improve care:


  • Increase awareness for the need to decrease radiation dose to children during CT scanning. Protocol development recommendations are offered under What can I do?
  • Be committed to make a change in your daily practice by working as a team with your technologists, physicist, referring doctors and parents to decrease the radiation dose! Sign the pledge! Click on the link on the home page to join the image gently campaign today.
  • Contact your physicist to review your adult CT protocols and then use the simple CT protocols on this website to “down-size” the protocols for kids. More is not better….adult size KV and mAs are not necessary for small bodies.
  • Single phase scans are usually adequate. Pre- and post contrast, and delayed CT scans rarely add additional information in children yet can double or triple the dose! Consider removing multi-phase scans from your daily protocols.
  • Scan only the indicated area. If a patient has a possible dermoid on ultrasound, there is rarely need to scan the entire abdomen and pelvis. “Child-size” the scan and only scan the area required to obtain the necessary information.



Imaging is expensive and growing rapidly. My guess is that CT utilization could be better controlled if patients had a better awareness of what’s involved.

January 24, 2008

11 thoughts on “Image gently, or when the diagnostic is worse than the disease”

  1. It would be helpful to have a simple statistic such as the chance of getting fatal cancer from a non-contrast CT scan. This would make it easier for doctors and patients to evaluate the risk, and add it to the usual considerations of monetary cost and the expected benefit.

  2. I think that radiologists should add a section to the consent form they have patients sign, prior to performing CT scans, that includes “increased risk of cancer from ionizing radiation”. Maybe patients, and providers, will think twice before using these tests in excess. There is an explosion in the use of diagnostic CT over the past decade. These scans are often helpful, easy to order and perform, and they are fun to read. As a result, we have many, many patients who receive dangerous amounts of ionizing radiation and very likely develop cancer simply as a result of these “diagnostic tests”.
    Given what we know at this point, medical providers need to seriously consider the risks involved in ordering CT scans. Many times, there is an alternative diagnostic modality (eg., ultrasound, MRI) that is entirely appropriate to r/o or rule in the pathology that is being considered. I have seen many patients who have had >5 CT angio’s in one year to r/o PE (they keep presenting to the ER with atypical CP and SOB). This is likely an un-intended side effect of the increasing use of hospitalists (as opposed to primary care/gen med providers who know their patients on a continuity basis) who have a tendency to look at every patient with a ‘fresh eye’ (eg., they rarely evaluate the patient in the context of more than a few weeks).

    What I love most, is the radiologists who read a study and then “recommend” three or four different studies to more clearly define some BS finding they are too chicken to actually “call”. (“3mm pulm nodule, recommend f/u CT, Small indeterminate liver lesions, recommend MRI, renal cysts, recommend U/S”). No wonder we spend so much on medical care in this country.

    Sorry for the rant…

  3. Last week, I saw a discussion on a medical blog about the risks of CT scans:

    Apparently, some time ago the doctor didn’t order a CT (which was really NOT indicated) and missed something. The question in court was “what would it hurt to have ordered it?”. The thread of comments answers this question – a pretty interesting reading. It also mentions other risks, in addition to radiation – potentially fatal allergy to dye; as well as risks present in all tests, not just CT scans – false positives, overdiagnosis.

    In general I think that while we all know about benefits of finding something early, most of us know very little of risks associated with overtesting. I think we all would benefit if both benefits and risks were explained to us clearly without overemphasizing the benefits and sugarcoating the risks. Also, the benefits should be expressed in absolute rather than relative numbers. “Decrease the risk by X%” is totally meaningless unless one knows what one’s risk is. Yet, “decrease the risk by 33%” sounds so much more impressive than “decrease the risk from 3% to 2%”. To be able to compare the probability of benefit with probability of harm, we need to know absolute chance of both.

Leave a Reply

Your email address will not be published. Required fields are marked *