A new study, published earlier this year in The Lancet, presented real patient data on the risks of cancer from CT scans in kids (Lancet. 2012;380:499-505). Though small, the risks were statistically significant and showed a relationship with dose. The study received quite a bit of mainstream news coverage and raised the issue, once again, of how best to weigh the benefits of CT scans against their risks, especially in pediatric patients.
Explore this issue:September 2012
Concerns about the increasing use of medical imaging and the concomitant risks of radiation exposure have been around for some time. In 2008, the World Health Organization launched an initiative to decrease unnecessary exposure. In 2010, the U.S. Food and Drug Administration began its own initiative to promote safe use of CT scans (and other medical imaging tests), including efforts to improve clinical decision-making and increase patient awareness.
One certainty is that overall use of CT scans is on the increase. A 2009 study found a three-fold increase in the number of CT scans run over a 15-year period, culminating in 72 million CT scans in 2007 (Arch Intern Med. 2009;169(22):2071-2077). About 7 percent of those scans were run in children. Not only are children more sensitive to the damaging effects of radiation, but they also have a longer potential life span in which that damage can morph into cancer.
How much otolaryngology has contributed to the increased number of tests is not precisely known. It’s also unclear how the availability of in-office CT scanners—and the potential reimbursability of their use—may influence an individual physician’s practice.
Despite all this concern about medical radiation, quantifying potential cancer risks has been largely a theoretical exercise to date. Using data from the biggest doses of radiation ever given—the atomic bombs dropped on Japan in World War II—scientists have tried to estimate dose-related risks based on incidence of cancers developed after the bombings and patients’ distance from the impact site.
“People have always been concerned but have never been able to collect actual data,” said Alan Craft, MD, emeritus professor of child health at Newcastle University in the U.K. and one of the authors of the Lancet study, which he calls the first to have enough CT scans and enough cancers to do the analysis. “The results are not surprising. They’re telling us what we thought.”
The Lancet study examined every person younger than age 21 who received a CT scan through the U.K.’s National Health Service between 1985 and 2002. These patients were cross-checked with U.K. cancer registries for leukemia and brain cancer. Doses absorbed by bone marrow and brain tissue were estimated for each type of CT scan given based on typical machine settings in young people and were adjusted over time, thus accounting for lower doses per test in later years of the study window.
Dr. Craft said that kids with the highest exposures—at least 30 mGy—had a risk of developing cancer that is three to five times greater than those with the lowest exposure—less than 5 mGy. “In absolute numbers, the risk of leukemia went from 1 in 2,000 to 1 in 300 to 400,” he said. “It’s still incredibly rare. But it’s still clearly an increased risk.”
Obviously, the study was retrospective and observational, not a clinical trial. Some scientists point to other potential weaknesses, such as the lack of a no-exposure control group and no data as to why the CT scans were done. Also, said Robin Cotton, MD, director of otolaryngology at Cincinnati Children’s Hospital Medical Center, “Doses have been going down, down, down.”
Some have maintained that nothing untoward happens with the much lower doses of radiation received in medical imaging, said Mike Hanley, MD, an attending radiologist at the University of Virginia School of Medicine in Charlottesville. At the very least, the Lancet study, which used data from medical records in the UK, suggests that there’s some risk at lower levels of exposure, he said.
The risk is real but very tiny—at least for a single scan. “I don’t think much about one CT scan,” Dr. Cotton said. “But I do think about two CT scans.”
On the plus side, sinus CT scans deliver relatively low radiation doses (typically less than 1 mSv) compared with chest (6-7 mSv) or coronary CT scans (12-16 mSv). But that shouldn’t give otolaryngologists license to scan without pause. Of course CT scans should be administered only when warranted, Dr. Craft said. Still, “people have got to think twice and thrice. Each CT scan adds to risk,” he added.
The benefits of CT scans for diagnosing sinus disease are well known to ENT surgeons. By the time patients are referred, they probably have already been through some diagnostic tests and medical treatments. Surgeons use the CT scan to provide a quick, definitive answer and to determine who might benefit from surgery, said Terence Davidson MD, an otolaryngologist/ENT surgeon at the University of California San Diego School of Medicine. The test is also used to provide a roadmap for sinus surgery, said Dr. Cotton.
Physician Knowledge Sometimes Lacking
There are no clinical practice guidelines regarding when to use sinus CT scans in adults or kids. The burden is on physicians to understand the radiation exposure they’re giving, as well as associated risks. Patients are asking for this information more frequently, in part because of the news generated by the Lancet study, as well as a recent study about dental X-rays and brain tumors (Cancer. 10 April 2012, doi: 10.1002/cncr.26625).
When patients began asking Cmdr. Alexander Stewart, MD, an attending physician in otolaryngology head and neck surgery at San Diego’s Naval Medical Center, about risks from CT scan radiation, he realized he didn’t know enough to answer with authority. “What’s the dose I’m recommending? What’s the risk? I really didn’t know,” he admitted.
As he tackled the literature to learn more, Dr. Stewart learned he wasn’t alone. A review of physician knowledge about CT radiation doses and risks showed the majority of doctors surveyed were on shaky ground (Eur J Radiol. 2010;76(1);36-41). He also read about variability in doses with the same type of scan and decided to do his own study. “The impetus was to educate specialists, our ENTs, about it,” he said.
Dr. Stewart worked with a radiologist and a radiological physicist to take measurements from a variety of in-office, hospital-based and outpatient imaging facility CT scanners used by otolaryngologists for sinus scans and found a 10-fold difference in radiation doses delivered (Int Forum Allergy Rhinol. 2011;1(4):313-318). The machines themselves were not the main factor in the variability measured, he said. “It’s the input into the machine—the protocols—that’s what’s variable.” Further, he said, manufacturer’s guidelines are typically optimized for the best image, not for keeping radiation doses as low as reasonably achievable—ALARA.
Legal regulations to report and track radiation exposure from medical tests may also come into play. In California, for example, a new law requires the reporting of radiation dose for every procedure.
What Can Practitioners Do?
In addition to educating themselves and taking the time to talk through the risks and benefits of CT scans with patients and their parents, otolaryngologists should collaborate with their radiologist colleagues. “They’re your local experts on radiation safety and doses. They’re the most familiar with [scanner] protocols,” Dr. Stewart said.
Lower doses are indicated for children. Dr. Hanley said some machines have a “pediatric button,” which is too nonspecific to be useful. “There’s a big difference between a 2-year-old and a 16-year-old,” he said. “One size doesn’t fit all.”
Other local resources include hospital radiation physicists, who are responsible for quality assurance, and even the technologist who works the scanner. “A lot of things can be done at the technologist level,” said Dr. Hanley.
Scanner manufacturers are working on the issue, Dr. Hanley added. “From the industry side, all the buzz is how to reduce dose.” The two main strategies used are scanning faster and boosting the software’s ability to reconstruct images.
In addition, maybe especially with the youngest patients, physicians should pay attention to cumulative dose. “I do it now. I’ll look and see how many scans someone has had,” said Dr. Stewart. “I take that into consideration.”