Session I: Helical CT and Cancer Risk

Panel discussion

Comment, Archie Bleyer, Children's Oncology Group: I want to thank everybody for inviting us and putting on what's been a remarkable high-impact source of information. With that, I would like to ask three questions.

I didn't hear enough about the latency of cancer. I know it took 50 years to get the data we finally needed to show that there is a dose-response relationships at even the very lowest doses, and so I assume it takes a very long time to get these cancers. Is the latency of the cancers the reason it took so long to conclude that low doses are carcinogenic?. How long does it take? Is it 10, 20, 30 years or longer, and what kinds of cancers are we expecting to see at the earliest? Types of leukemia were mentioned. Is this a problem for young people or is it a problem when they reach middle age or are even older?

Dr. Elaine Ron: It didn't take 50 years to see a dose response, it took 50 years to see what the lifetime radiation effects are. We saw increases in radiation risks fairly soon after the A-bomb. Leukemia has the earliest latency. Generally, you can see increased risks for leukemia by 2-5 years, but it persists. For solid cancers, it's generally longer and usually 10-20 years, but it depends on the cancer. For thyroid cancer you can see it much earlier. Even in external radiation, you start seeing it in about 5 years. For some of the other cancers it may be 10 or 20 years and it persists through life. For the children, it depends on what age they are irradiated, but certainly for those who have been treated with total-body irradiation for Hodgkin's disease, you see it fairly soon, so it is a concern.

Dr. Archie Bleyer: I think this issue is relevant to the question of what kind of study we would design if we wanted to evaluate the cancer risk of recent and current scanning practice. We would certainly have to plan a leukemia study and not a solid cancer study given we're less than 10 years into the spiral CT era.

Dr. Elaine Ron: You might get an answer in 5-10 years if you looked at leukemia specifically.

Question: The next question is about the sex or gender difference. What are the biological explanations for females being at much higher risk? Double was the estimate thus far over males. Is it the excess in breast cancer or is it the differential growth rate, because females grow faster in the first 10 years? What is the biological explanation thus far?

Dr. Elaine Ron: I can give a few tentative answers. First of all, there is a difference in spontaneous cancer rates for men and women. In fact, men have higher rates than women. In terms of the radiation induced, it depends partially on the cancers; and again, thyroid cancer and breast cancer are obviously much more common in women and that does somewhat increase the women's rate. In addition to that, some people feel that there might be some interaction with hormones. Another possibility is smoking - the high rate of smoking in men makes them get smoking-related lung cancers. Women have lower rates of lung cancer and therefore they can get other types of cancers and the radiation can have more of an effect.

Question, Dr. Nancy Rosen: My question is for Dr. Brenner. You've talked to us about estimated risks at mAs of 300 and 200, you didn't mention the kV or the pitch used. As Dr. Frush will discuss later in the program, many of us have used reduced dose scanning for many years and it's possible to reduce the dose by other ways than by just decreasing the mA. It has been pointed out that reducing kV and increasing pitch lowers dose. Could you give us the estimates of the risk using the doses that are recommended using lower kV, lower mA, and increased pitch? And is it possible then that, if we lower things enough, there would be no risk?

Dr. David Brenner: Average kVs are used. I can't see why, if you lowered the dose, you'd ever get to a point where the risk is 0. There is no evidence that I can see that there is some threshold below which the risk is 0.

Dr. Nancy Rosen: But might it be so low that we couldn't measure it? Or could you just tell us before the meeting is over what the risk would be?

Dr. David Brenner: If you half the dose you'd half the risk. It is certainly true that if you reduce the dose enough, you would get to a risk that you could never measure because you'd be in the noise. That doesn't mean the risk is 0 because you can't measure it. Obviously our objective is to reduce the dose as much as possible and thus comparably reduce the risk by that same factor.

Dr. Nancy Rosen: Yes, that is everyone's objective. Thank you.

Dr. David Brenner: I don't see that's evidence that you would ever reduce the risk to 0.

Question: I was aware of this paper from Canada about one or two X-rays and the increased incidence of leukemia, which was startling to read. I'm sure it includes virtually no CT data because of the lack of CT scanners in general in Canada, and the time when it came out. I wanted to tell you that I've seen three children in the last 2 years with osteogenesis imperfecta who have developed leukemia. I'm curious if there are any epidemiologic data on this. It's not a terribly common disease, but it is a reasonably common disease in major medical centers where there are clinics. These children have hundreds of X-rays. Do we have any data on the incidences of leukemia in that disease?

Dr. Elaine Ron: I cannot think of any specific study, I would have to go back and look at all the studies to see if they have ever broken that out separately. But I am not aware of anything.

Comment: I don't know of any reported relationship between osteogenic imperfecta and cancer, including leukemia. Biologically, I would worry about that because of the relationship between bone, developing bone, the marrow, the communication between the microenvironment and the known progenitors that ultimately lead to leukemia and other malignancies in the system. I would think there might be a relationship. But it hasn't been reported that I know of. You may have just done that.

Comment: I would also like to thank all the panelists for their balanced presentations of a very difficult topic. From my point of view as a practicing pediatric radiologist, I have to remark that I've often seen the scale with the risks and the benefits ratio, with the risk being high up and benefits way down. As a radiologists it is not always clear to me in individual cases that the benefits are truly higher than the risks. I've seen many requests for trivial indications. There is a tremendous difference between various countries (e.g., USA and UK) in utilization. I think it's a very good thing to explore why these differences exist and what can be done about it. The other issue is appendicitis. It's almost assumed that ultrasound is inferior, which is shown by studies. But it doesn't necessarily mean that ultrasound should not be used as screening tool.

Comment: In terms of reducing the collective dose to the population, it seems to me that you are going to have a greater effect by reducing the average dose for all CTs than saying perhaps there are a few CTs we don't need to do. Certainly one should do both. There is no reason not to do both. In terms of the overall public health benefit, I think reducing the average mAs and perhaps pitch settings is going to have a bigger benefit than looking at particular studies that don't have to be done. I don't know if everyone else would agree with that.

Dr. Tom Slovis: I don't agree with that! I think we could ask the pediatric radiologists in the room, "If you had to pick a number, what percentage of CTs being done, either head or body, would you think they could probably do without?" If I were to pick a number, I would say the lowest number would be 10% and maybe considerably more than that. Would the pediatric radiologists agree with that? (The audience thought up to 30% of studies were not necessary.) I think clearly we can reduce radiation the most if we can get rid of the unnecessary exams. That's the number one thing we need to do. We also desperately need to do all the things you are talking about. If 10% of CT exams could be avoided, you would presumably reduce the risk by 10%. If you could halve the dose, you would presumably reduce the risk by 50%.

Comment: It is quite clear that there is an interaction between both of these things. Where you get the biggest bang for your buck can be a matter of controversy. Clearly we have got to do both. It's going to depend upon the environment we are in. Many of us are university radiologists or university physicians. We have a very different environment than other hospitals and other situations and don't know what is going on in private practice and in other hospitals and other situations. I think you can get large benefits from both of these issues and the question that I always like to ask is, "When we use the radiation or we find a new way of using it, are we getting the necessary increased benefits for the added risks that we are imposing?" It's just like a new drug, it's like that kind of benefit:risk ratio. We do have to be self-critical about our usage of radiation in this manner.

Comment, Edward Staab, MD, National Cancer Institute: I wonder if anyone could give me some basic data: Of all the CT scans done in children, how many are supervised by pediatric radiologists? How many are done by adult radiologists? And in a similar fashion, how many CT scanners are located in hospitals, in academic centers, nonacademic centers, and non-hospital environments? And finally, how many scanners are located where radiologists are not supervising them at all? (No answers.)

Dr. Walter Berdon: Since this is a big friendly group, I will speak frankly. I watch scanners going into community hospitals where friends of mine work, and know about the interchange between the companies putting the scanner in and the technicians and occasionally the doctors. There is no physicist involved early in the procedure, and maybe not even at the end, to check it out. I think it is our responsibility to embark on a broad educational campaign. The companies share an equal responsibility in this situation.

The companies have such an enormous role in the installation of these machines. In New York City, there is an "outbreak" of scanners outside of hospitals. These are very busy, doing total-body scans, cardiac CTs, and are staffed by radiologists making a lot of money, but there is no supervision of dosimetry. This is where adult radiology is going and I don't know where pediatric radiology is going. Probably not many children are being directed to those scanners. But right now it is totally out of control and at some point we have to get together with the companies and the radiologists and start talking. I think that most radiologists I know have never seen a radiobiologist except during their residency.

Comment: I liked your comment about the community hospitals. I recently did an inspection of a community facility, an outpatient facility. I found that for every CT scan they performed, they drove the machine to the maximum. They did it at maximum technique every time they put the patients through. Highest dose all the way. They didn't even know what they were supposed to do. The head of the area thought that they were doing something entirely different. They've been working this way for 2 years. Your comment about the fact that there is no supervision is well taken. I would also like to comment that you've brought up the money factor in this benefit:risk issue, which I think is a very important issue and a tremendous driving force. How we get in control of it, I don't know.

Dr. Janet Strife: I would like to address the responsibility of the manufacturers and the board-certified radiologist. I personally think it's misdirected to look at the manufacturer of equipment to see what's happening with radiation doses in children and adults.

Dr. Slovis: I would like to thank our panel, they did a terrific job. We really need each other and I think it became very obvious today that we don't read the same literature. Many, if not most, of those quotes today from articles from Canada and Shanghai are in journals that the radiologist doesn't usually see. We desperately need each other and we need to keep this dialogue going. Bill Radaj said last night and it came out again today, the training of people who run our equipment and our training when we get new equipment may not be very good. I would use another expression, but this is being audiotaped. We really need to have more information coming both ways. We have to ask for it and the manufacturers have to provide it; and we have to make sure that our techs are in fact qualified in a way that we want them to be qualified and certified where there is certification. There is a lot to do. I can't agree more with Janet, though; the buck stops with us as board-certified radiologists. We are diagnosing the patients and we need to know what the dose is and we need to know what we are doing.