The Doctor Says Get a CT Scan. Should You?

A new report in tomorrow's issue of the New England Journal of Medicine raises serious concerns about the use, and overuse, of CT scanning. While individual risks of developing cancer from a CT scan, which emits high doses of radiation, are relatively low, the researchers worry that their rapid growth as a highly accurate diagnostic tool is exposing too much of the population—and an increasing amount of vulnerable children—to radiation and might be setting the stage for higher incidence of cancer in years to come. Around 62 million scans are performed per year, compared with only 3 million in 1980. Moreover, the researchers estimate that a third of those CT scans are entirely unnecessary—many of them now performed by cautious doctors on worried people with no symptoms at all.

How can you know if a CT scan is a wise move? U.S.News & World Report spoke with Fred Mettler, chief of radiology and nuclear medicine at New Mexico Veterans Health Care System and an expert on the effects of radiation, to find out.

How is a CT scan different from a traditional X-ray?

With a traditional X-ray—a chest X-ray, for example—radiation goes through you from one side to the other, with 3-D information ultimately projected onto a two-dimensional picture. With a CT scan, an X-ray tube rotates around the patient and presents the results to you as a three-dimensional picture. The advantage is that it's much more sensitive, is high resolution, and offers much more anatomically specific information with great detail.

But a typical chest CT means around a 175-times-greater dose of radiation than a similar chest X-ray—that's like 20 sets of mammograms. That range can be between 20 and 200 times higher depending on which part of the body is being scanned.

When is a CT scan definitely warranted?

If you think something is desperately wrong with you and you require an immediate answer. For example, an incredible abdominal pain or a severe acute headache as opposed to a migraine you have had off and on for years. If the CT scan is medically needed, don't think about radiation. If you have excruciating head or abdominal pain, radiation exposure should not be high on the list of concerns.

But there's no question that it's overused—and I don't think I can quarrel with the number of 30 percent.

How should judgment factor in?

One common problem for adults is renal stones. They have them all the time on and off, typically with horrible pain. It used to be that we gave them water and pain medication and told them to wait for the pain to pass. Now, all urologists want a CT scan done. I've seen people in their early 30s who have had 18 or 20 CT scans. They come in once a month, but there's no evidence in the literature that this sort of thing is justified.

Another good example is if you have significant abdominal pain and go to the emergency room. The physician in the ER is buried with a "bazillion" patients. He can poke on your abdomen and see where the pain is emanating from, thinking it's appendicitis. Then he'll get some blood tests and if you have a high count will call a surgeon. Before the use of CT, around 20 percent of the time you got operated on and had a normal appendix.

Nowadays, doctors will immediately get a CT scan, and because it's so sensitive and accurate for diagnosing appendicitis, less than 1 or 2 percent end up taking out normal appendixes. An additional advantage is that CT allows the doctor to see the entire abdomen and pelvis, the aorta, kidneys, and gallbladder. If your doctor wasn't in fact right about the appendicitis, the CT scan will pick up a lot of other things that might be giving you your pain. It's incredibly efficient and, of course, doctors can move you through ER much more quickly.

Is using a CT to determine appendicitis generally justified? Well, if the patient has severe pain in the right lower quadrant, then probably yes. On the other hand, is it justified for a little kid who is crying and doesn't know what's going on? In that case, probably not.

What are the risks involved in CT scanning?

Certainly a CT scan won't make your hair fall out or anything so drastic. But the risk of cancer is there, and it depends on a few things. There's the dose of the radiation, which depends substantially on the age of the patient. We know that kids are more sensitive and their risk of getting cancer is higher [when they are given] an adult dose. But these cancers tend not to occur for years or decades at least. For children, the risk of developing a fatal cancer is somewhere around 1 in 500 or 1 in 1,000—the older you get, the lower the risk becomes. So if someone is 90, for instance, there is virtually no cancer risk. For an adult, the risk is around 1 in 2,000.

At this point, people believe there is [a] linear relationship between the dose and the risk of cancer. So if you cut the dose in half, you also cut the risk in half. The question should be, can you optimize the dose by using a lower amount of radiation or can the scan be done only once instead of three times? For kids especially, doctors should really optimize with the lowest dose possible if a CT scan is warranted.

Is that happening?

I would say that five or so years ago, there's no question they were actually getting adult doses. But after Dave Brenner [of Columbia] started hopping up and down about this problem and it was picked up in the media, pediatricians got very sensitive and manufacturers began putting pediatric dose settings on the CT machines. But, I don't think the sensitivity is there for teenagers or young adults.

What are some alternatives to CT scanning that patients should know about or ask their doctors?

I would ask if there are ways we can find out the answer without using radiation. Using ultrasound, for example, or doing an MRI scan. Those don't use ionizing radiation, so there is virtually no risk. And if the scan is definitely going to be done, ask about the dose you will be getting. If a child is getting a scan, a concerned parent should make sure the radiology technician is using the correct pediatric doses.

How do you prevent redundant CT scans and help patients to keep track of the scans they've already had?

When our patients go to another hospital or doctor, we give them a CD of all their exams on it, and anyone can open them up, throw it on a PC, and have a look at the pics. That's very helpful.

Are CT scans useful for asymptomatic patients?

I don't recommend CT scans for asymptomatic patients. Take lung cancer screening for example. One article in the New England Journal of Medicine from last year concluded that CT scanning could prevent 80 percent of lung cancer deaths by catching lung nodules. Then in a 2007 Journal of the American Medical Association study, there was found to be no reduction in mortality by using CT scans for lung cancer. We shouldn't do something with risk until studies determine what the risks and benefits actually are. A lot of scans are not justified, like the business of whole-body CT screening. No one has ever shown that these have more benefit than risk.