Radiation Exposure and CT Exams

By Joyce Graff, Director of Wellness at the VHL Family Alliance, vhl.org

I was privileged to attend the conference “Management of Radiation Dose in Computerized Tomography: Toward the Sub-mSv Exam”, held by the Coalition for Imaging and Bioengineering Research (CIBR) in Bethesda, Maryland, February 24–25, 2011. Most of the 140 attendees were radiologists and medical physicists from major medical centers, scientists from imaging manufacturers, and representatives from regulatory agencies (FDA, NIST, etc.). I was one of five consumer advocates at the meeting.

The goal of the conference was to discuss ways to reduce the total radiation dose required during one CT examination to below 1 milli-Sievert of radiation per examination. The group was earnestly focused on providing “the right scan to the right patient at the right dose.” Please note that there was no discussion at this conference of the use of ultrasound or MRI—this group was entirely focused on optimizing the use of CT technology.

First, what’s a Sievert? The measure of the radiation dose in milli-Sieverts (abbreviated mSv) was defined in 1979 to allow physicists to compare the amount of energy deposited in the human body by exposure to ionizing radiation, across machines and methodologies. It is named for the Swedish physicist Rolf Sievert (1896–1966), one of the pioneers of medical physics. http://www.sizes.com/units/sievert.htm.

To put that into perspective, background radiation (walking around in your city) gives you about 3 mSv per year. A flight from New York to San Francisco delivers 0.03 mSv.

We have long asked how all that radiation exposure may be cumulating, adding up, and potentially affecting our long-term health.

Why, one might ask, would I even sign up for such a conference? As a patient advocate, I represent people who depend upon medical imaging to manage their health.

In the “old days,” before medical imaging was available, in order to see how big a tumor was or how threatening, one had only one choice: do “exploratory surgery”, remove the mass, put it under the microscope, and decide. Much of the time the tumor was benign, other times it was malignant, or when it was hard to make that decision in the operating room, the surgeon removed more rather than less to be on the “safe side.” This resulted in a great deal of excess surgery (e.g., to remove a benign tumor), subjecting the patient to pain, suffering, infection, and general wear and tear. In the VHL community, where people deal with multiple tumors over the course of their lifetime, this took an awful toll. The surgery was often clearly necessary, but would have had an even better outcome if we had known enough to do it earlier, before it reached crisis proportions.

Using medical imaging, we are now able to track the development of tumors, watch their threat level, analyze their nature with relative confidence, and “choose the optimal moment” for interventions, remove tumors when necessary, stretch the intervals between surgeries, and most of the time avoid surgeries for benign conditions. That does mean that people in our community may have had 1–4 CT’s every year for decades. We have long asked how all that radiation exposure may be cumulating, adding up, and potentially affecting our long-term health. The answer, even at the end of this conference, is: “we don’t know.”
It was a fascinating two days. Most of the details, of course, went far over my head, but I learned a great deal. The following are my observations as a patient and as a patient advocate.

The amount of radiation the human is exposed to in an examination is a function of

• The machine
• The operator and/or hospital protocol being used
• The organ to be studied
• The density of bone on the path to the organ under study
• The amount of adipose tissue on the path to the organ (the patient’s body-mass index, or BMI)

The machine vendors are definitely working on it, and the latest machines from several vendors go a long way to achieving this goal. BUT the very latest round of machines, available in Europe since 2009, are still not available for sale in the United States. Why? They have not yet been approved by the U.S. FDA.

These machines have buttons, knobs, and intricate settings in place to serve the needs of academic institutions with one or more medical physicists on staff. All parameters can be adjusted to meet the needs of any possible situation. As a result, the settings are complex, and vary considerably from one model to another within a manufacturer, and even more widely across vendors.

In an environment like the Mayo Clinic, Massachu­setts General Hospital, Memorial Sloan Kettering, or University of Michigan, there are 10–14 machines from 2 or more manufacturers, and of varying ages. The machines are expensive and are not “turned over” quickly. But of course in these places there are also 4–6 Ph.D. medical physicists on staff. To guide their imaging technicians, Dr. Cynthia McCullough of the Mayo Clinic showed a chart they had worked out that tells the technician what settings to use to image a particular organ on each of their machines. At Mass General there are over 300 such protocols for different diagnostic examinations.

But what about centers where there is not sufficient staff time and/or talent to work this all out? Furthermore, as one physician explained to me, what about the community hospital who has their part-time day technician, their part-time night technician, and the fill-in guy who usually works at another hospital (on a different machine) and fills in when necessary?
Dr. James Thrall of the Massachusetts General Hospital explored in depth the challenges of reducing the radiation dosage. Years ago, the average dosage for one CT exam was in the range of 20–25 mSv. The average dose per exam today is about 7 mSv (or 14 mSv for abdominal CT). To achieve the goal of less than 1 mSv (which is approximately what you get with three chest x-rays), the picture quality suffers. He showed pictures comparing a low-dose study today with the higher-dose studies of 20 years ago, which were in fact quite comparable. The machine improvements have improved image quality enormously at the dosages previously used, but reducing the dose today brings us back to the lesser quality seen in the older machines. “But we read them then, and we can learn to read them today,” he said, distinguishing the gray-on-gray subtleties.

From a patient’s perspective, that means we are dependent upon the talent and experience of the radiologist. Reading a medical image is an art, not a science. And finding an artist at a community hospital is again very difficult to do. Even getting matching readings from two talented radiologists at major medical centers is very difficult. I have experienced more than once that a patient sending copies of the exact same scan to multiple radiologists will get back entirely different readings. The number of tumors, the size of tumors, and the nature of the mass may all be read differently based on the experience of the radiologist with this particular tumor type.

Until recently, at some centers, dosages calculated for adult patients were being used also on pediatric patients. Obviously the body-mass index for a 2-year-old is considerably different than for a 6-foot 250 pound adult. There is a nationwide campaign, called “Image Gently” under the direction of Dr. Marilyn Goske of Cincinnati Chidren’s Hospital, to educate radiologists about the different calculations needed to ensure that children are not subjected to too much radiation. As Dr. McCullough said, “from the neonate to the morbidly obese, the safe dosages can vary by a factor of 20.” We don’t need “pretty pictures” for all diagnostic tasks. Is the image sufficient to answer the presenting question?

As patients, we are often asked to repeat the same imaging study because the doctor from whom we are seeking a second opinion either cannot get access to the study done at the previous hospital, or doesn’t like the image quality from that other machine. How much better it would be if there were more consistency in images from one center to another!

I raised the issue of obtaining and viewing images from one center to another. Even when you carry your CD from one doctor to the next, there are instances where the neurosurgeon spends half an hour fussing with the readability of the images on his or her computer. It would be great if there were a portable format, like a “pdf” for medical images that would allow scans to be more widely read. Evidently, there is an effort underway at the NIH to achieve this.

For those of us who have had repeated images for years, the dosages are rarely captured in the medical record. In order to study the long-term effects of radiation-based imaging, we will need to begin to capture in all electronic health records the human dose data associated with each scan. This would enable retrospective analysis of the long-term effects.

Certain genetic variations make some people more sensitive to radiation exposure than others.

Certain genetic variations make some people more sensitive to radiation exposure than others.People with ataxia-telangiectasia, for example, are very sensitive to radiation. There is an effort to produce “radioprotectors” especially for people getting radiation treatments, but for ultra-sensitive patients they might also be used for diagnostic studies. In Kerala, India, for example, where there is particularly high background radiation, adding a dose of diagnostic imaging radiation has higher impact than in other regions.

There is definitely progress being made. In cardiac imaging, several centers have already achieved the sub-mSv CT exam which does provide the requisite information, and there has also been an effort to reduce the frequency of screening wherever possible. But that knowledge is confined to the major centers and has not trickled out to most of the community hospitals. As one of the panelists said in the wrap-up session, there is still a wide “knowledge gap between the people in this room and the practicing radiologists in the field.”
As Dr. Thrall outlined in his keynote address, what it will take to achieve these goals includes:

• Regulations and national standards
• Improvements in the process of FDA approval of new machine models
• Education
• Improvements in the culture of clinical practice
• Reference dosages
• Registries and tracking of dose-per-exam to confirm dosages and to allow retrospective studies.

“Medicine is a team sport,” he said. “This industry touches every hospital. Dose optimization and minimization will require a blend of hardware and software improvements, and policy and practice… NOW.”

And let’s not forget that the patient is a member of this team, weighing the benefit/risk ratio for each scan. Before the patient even gets to the Radiology Department there are two additional team members who are making decisions. We need to be able to articulate benefit/risk in terms that are understand­able to the patient and to the ordering physician so that they can make reasoned decisions like:

• What are we trying to accomplish?
• Which kind of scan is best in this situation?
• Do I need contrast media? Do I need a scan before and after contrast (doubling the radiation dose)?
• What is the exposure to radiation?
• What will it tell us?
• Is there another medical test that will give us equivalent information?
• Can I afford this?
• Will my insurance company pay for it?
• Can I afford the time away from work or family?
• Am I willing to put myself into that tunnel?
• My child breaks into tears at the thought of yet another scan. Can I do this to him? How important is it really?

With the increasing availability of Magnetic Resonance Imaging (MRI), we should encourage the use of MRI rather than CT wherever possible. This is a discussion which has to happen in the office of the prescribing physician, not in the Radiology Department. I would encourage patients to engage their physician in this discussion. As the decision is made among x-ray, ultrasound, CT, and MRI, the focus should be on how best to answer the diagnostic question while minimizing exposure to radiation—and that should include a serious look at whether an MRI might not be a better choice.

As printed in the VHL Family Forum 19:2, April/May 2011. For permission to reprint, please contact VHL Family Alliance, editor@vhl.org. Further information is available from the VHL Family Alliance, info@vhl.org.