Potential link between radiation exposure, cancer risk still debated
Recent developments bring medical imaging into the focus of health care discussion.
Recent positive studies and some negative publicity have brought the subject of CT scanning to the forefront of discussions within the health care community. The main issues at hand are the frequency of exposure and the radiation dose administered to individual patients on scan-by-scan bases, and how those individual risks add up at the population level.
For instance, The Data and Safety Monitoring Board for the NCI’s National Lung Screening Trial recommended that the trial be halted because of statistically convincing findings that low-dose CT scanning provided a clear mortality benefit among older heavy smokers.
Shortly thereafter, the American College of Radiology launched the Image Wisely campaign, which aims to raise awareness of the benefits and risks of medical imaging, particularly with regard to cancer risks linked to radiation exposure.
Photo courtesy of:
These events occurred after the largely negative exposure that Cedars-Sinai Medical Center received last year, when approximately 385 patients from six hospitals were exposed to excess radiation during CT brain perfusion scans, according to FDA sources.
Other areas of debate include the identification of patients most likely to benefit from CT scanning, the reliability of scan results and the effect of heightened attention on scanning use moving forward.
“The major message is that with any imaging procedure that uses radiation, there is a small risk that the radiation itself may cause cancer,” Christine D. Berg, MD, chief of the early detection research group in the division of cancer prevention at the NCI, told HemOnc Today. “It is helpful to understand to the best of our abilities what that risk is, and we want to be sure that the reason we are doing that imaging is such that the benefit will outweigh the small risk.”
The National Lung Screening Trial involved more than 53,454 current and former smokers, who had quit within 15 years, aged 55 to 74 years with a minimum smoking history of 30 pack-years but who had no signs or symptoms of lung cancer. The trial took place at 33 sites nationwide, officially opening in September 2002 and scheduled to close December 2009.
The researchers compared two lung cancer screening methods: low-dose helical CT scanning (LDCT) and standard chest X-ray. Participants were screened at enrollment and at the end of the first and second years of the trial. They were then followed up from between 2 years and 7 years. Deaths, particularly those verified to have been caused by lung cancer, were documented.
As of Oct. 20, when the review board conducted its final meeting, there were 354 deaths from lung cancer in the LDCT arm and 442 deaths in the X-ray arm, a reduction of 20.3% for LDCT scanning. Other findings indicated a 7% lower all-cause mortality rate in the LDCT arm compared with the X-ray arm.
Munir Ghesani, MD, an attending radiologist at St. Luke’s-Roosevelt Hospital Center as well as Beth Israel Medical Center New York, and associate clinical professor of radiology at Columbia University College of Physicians and Surgeons, said the findings are particularly significant in light of the ongoing publicity from the Cedars-Sinai incident.
“This is big news because there have been so many reports recently that looked into negative implications of CT scanning and radiation exposure,” Ghesani said. “That publicity has not been positive, but we must remember that there are still advantages to performing scans. The National Lung Screening Trial helped demonstrate that.”
Although the results are encouraging, William C. Black, MD, professor of radiology in the department of radiology at the Dartmouth-Hitchcock Medical Center and investigator in the National Lung Screening Trial, said acquiring any definitive answers surrounding imaging and cancer risk is difficult.
“We looked at the leading cancer, which is lung cancer,” Black said. “We know who gets lung cancers — people who are older than 55 and smoke a lot. We had the best possible scenario for screening, where we were basically looking for a white dot on a black background. But even under those ideal circumstances, we needed 53,400 people and 7 years of trial at a cost of about $250 million to answer the very simple question we posed.”
He said the trial answered only the question of mortality benefit of scanning in a specific high-risk population, but complete data, including potential risks, have not yet been released.
Most of what is known about radiation exposure and cancer risk comes from extensive follow-up of victims and the children of victims of the atomic bombs that were dropped on Hiroshima and Nagasaki 65 years ago in Japan and from people living at or near the site of the 1986 Chernobyl nuclear accident in Ukraine.
Bruce J. Hillman, MD, of the departments of radiology and public health sciences at the University of Virginia, said the associations between cancer risk and these disasters are compelling, but the low levels of exposure in a CT scan have never been proved to cause cancer of any kind.
“Though we understand that there may be risks associated with radiation exposure in a CT scan, I have reservations about extrapolating back from Hiroshima,” Hillman said.
“Even with data from more than 53,400 patients, we still do not know if this low of a radiation dose from a helical CT scan will cause cancer,” Berg said. “You need a sample size of about a million for a dose that low, so it is all extrapolation at this point. Some people are uncomfortable with the level of extrapolation needed to develop this evidence, and rightfully so.”
According to Hillman, the challenge is to determine the boundaries of where the benefits will clearly outweigh the risks. He said the necessity of imaging is easy to define in the two extremes: Medical imaging is usually not necessary when there is little chance that a patient has a condition or when a diagnosis is all but assured.
“Medical imaging is best used in the middle ground, where there is a moderate amount of clinical uncertainty,” he said. “The problem is that there is no clear-cut point to where this mid-range lies.”
Berrington de Gonzalez and colleagues used modeling to estimate the risk for radiation-induced lung cancer mortality from three annual low-dose CT scans among those younger than 55 years and determine the benefits required to outweigh those risks.
For never-smokers, the estimated excess lifetime risk for radiation-induced lung cancer mortality from annual screening between ages 40 and 42 years was one in 10,000 (90% credibility interval, 0.4-3) for men and three in 10,000 (90% credibility interval, 2-6) for women. Risks were approximately twofold higher in current smokers.
The mortality reduction required to outweigh the potential risks ranged from 125% for female never-smokers to 2% for male current smokers aged 50 to 52 years. “Before age 50, the mortality reduction from lung LDCT screening that is required to outweigh the radiation risk may be substantial and, in some cases, unattainable (ie, >100%),” the researchers wrote.
Other studies involving CT scanning and cancer risk have resulted in similarly broad-ranging and occasionally inconclusive data, which has raised the necessity of a forum in which physicians and patients can stay informed.
Image Wisely campaign
The American College of Radiology, the Radiological Society of North America, the American Association of Physicists in Medicine, and the American Society of Radiologic Technologists collaborated to form the Joint Task Force on Adult Radiation Protection, which in turn developed the Image Wisely campaign.
The objectives of the campaign are to lower the amount of radiation used in medically necessary imaging studies and eliminate unnecessary procedures.
James A. Brink, MD, professor and chair of the department of diagnostic radiology at the Yale University School of Medicine and co-chair of the Image Wisely campaign, said the campaign’s website will serve as a repository of information to establish best practice. The site also contains a pledge that he urged imaging professionals to take:
- To put patients’ safety, health and welfare first, by optimizing imaging examinations to use only the radiation necessary to produce diagnostic quality images
- To convey the principles of the Image Wisely program to the imaging team to ensure that facilities optimize the use of radiation when imaging patients
- To communicate optimal patient imaging strategies to referring physicians and to be available for consultation
- To routinely review imaging protocols to ensure that the least radiation necessary to acquire a diagnostic quality image is used for each examination
“We want to communicate optimal patient imaging strategies, but we don’t want to throw the baby out with the bath water,” Brink said. “We need to recognize benefit, and that it has not been proven that CT scanning causes cancer.”
Population vs. individual risks
Brink said the lifetime risk for developing a fatal cancer is approximately 20% to 25%, and it is hypothesized that a typical dose radiation in a CT scan carries a risk of about one in 2,000, extrapolating back from the atomic exposures.
“When you add the hypothesized risk to the a priori risk and multiply it by about 70 million exams per year in the United States, you get a more substantial risk at the population level,” he said. “While we do not yet know if it causes cancer directly, it behooves us to control it as best we can.”
Berrington de Gonzalez and colleagues conducted another study, partially in response to the more than threefold increase in CT scans between 1993 and 2007.
“Overall, we estimated that approximately 29,000 (95% uncertainty limits [UL], 15,000-45,000) future cancers could be related to CT scans performed in the US in 2007,” the researchers wrote. “One-third of the projected cancers were due to scans performed at the ages of 35 to 54 years compared with 15% due to scans performed at ages younger than 18 years, and 66% were in females.”
Although data such as these present challenges for interpretation on a macro level, Black said it is easier to determine risk at the population level than it is in an individual patient.
Hillman said certain types of tissue are more susceptible to radiation than others. “Fast-dividing tissue is more susceptible than stable tissue,” he said. “Young tissue is more susceptible than older tissue. Tissues in certain organs such as the testicles or breast are more susceptible. Clinicians should be aware of these factors when evaluating the patient and the site of the scan.”
Thomas Guerrero, MD, PhD, associate professor in the thoracic section of the division of radiation oncology at The University of Texas M.D. Anderson Cancer Center, said the age and medical history of the patient are also factors.
“In older patients, the dose is not as much of a concern because of the shorter expected life span,” Guerrero said. “Patients who already have had a cancer are likely going to receive radiation treatment, so dose may not be a critical concern there, either. In pediatric patients, you have the most risk, but also the most potential benefit.”
Rita F. Redberg, MD, MSc, professor at the University of California, San Francisco School of Medicine and editor of the Archives of Internal Medicine, said the potential risks to an individual patient go beyond radiation exposure.
“A medical image that yields a false negative carries the double risk of radiation exposure and a disease that goes untreated,” Redberg said. “A false positive resulting in an ‘incidentaloma’ can lead to unnecessary invasive testing and treatment. As clinicians, we need to consider this whole package of risks.”
She said the potential harms of radiation exposure may not affect the patient for years. Most clinicians are familiar with the American College of Radiology Appropriateness Criteria, but Hillman said many simply do not have time to conduct a thorough review of the literature to determine the risks associated with each particular case.
“Moreover, data — and the recommendations within those data — vary enormously,” Hillman said. “Some are truly helpful and evidence-based, but some (are) eminence-based. Often, the reputation of the researchers who have conducted the study carries as much weight as the findings.”
Berg said although there is uncertainty on the individual patient level, one area that clinicians can control is the amount of radiation in each particular scan.
There has been much debate on the issue of dose reduction, but Guerrero said the effective dose should be reduced by 40%, adding that the dose required for CT imaging is somewhat dependent on the reconstruction algorithm.
“For the past 30 years, the algorithm has been the standard,” he said. “Newer algorithms have been developed and used clinically for nuclear medicine. The new algorithms have been able to produce good-looking images with a low number of events.”
Although advances in technology may allow more consistently lower doses, studies such as the one conducted by Smith-Bindman and colleagues suggest that inconsistency has been the standard.
“Radiation doses from commonly performed diagnostic CT examinations are higher and more variable than generally quoted, highlighting the need for greater standardization across institutions,” the researchers wrote.
The results indicated that radiation doses varied from 2 mSv for a routine head CT scan to 31 mSv for a multiphase abdomen and pelvis CT scan. When viewed across health care facilities, the mean variation between the highest and lowest dose for each study type was 13-fold.
Guerrero said he approved of studies such as those conducted by Smith-Bindman for bringing attention to the issue of radiation exposure. “Every measure to reduce dose reduces risk,” he said.
Although most clinicians said they would agree with this in theory, what happens in practice is often a different story.
“Fear of malpractice plays a huge role in the use of medical imaging,” Black said. “If a patient might have something wrong, it is simply better safe than sorry. Imaging is overused, particularly in the [emergency department], because of the fear of malpractice.”
According to Hillman, physicians who own expensive imaging equipment are far more likely to use it. “The best way to make back your money for a high-priced machine is to use it, often regardless of whether it is warranted or not,” he said. “When you have a hammer, everything looks like a nail.”
Redberg said physicians are rewarded for active steps taken, rather than passive observation. “There is a huge financial incentive to do more rather than less,” she said. “The whole issue is complicated due to the reimbursement structure we have.”
“We are rewarded for doing more, and not necessarily for doing it well,” Black said.
Regarding information in the public domain, Ghesani said, “Some centers were not following accepted protocols, and this created attention. But in spite of this, we have to be positive. It is important to capitalize on any publicity surrounding CT scanning and cancer risk and use it to raise awareness.”
Although raised awareness is positive, Hillman said there are potential concerns when patients have seen the news. “There is a wealth of information and misinformation out there. Informed patients entering a discussion is good, but insisting on directing their own care is bad. We have to be careful that they do not apply pressure and end up getting inappropriate tests,” he said,
“Knowledge among patients is a good thing. However, the inevitable downside to patient awareness is that the expectations for treatment can become skewed,” Redberg said.
In all likelihood, CT scanning will remain near the top of health care news pages in the near future. Data from Cedars-Sinai are still emerging; the full results of the National Lung Screening Trial are forthcoming; and the Image Wisely campaign is likely to gather momentum.
Ghesani said he took the Image Wisely message to heart. “Medical imaging is a good modality, but one that should still be used with judgment.” – by Rob Volansky
For more information:
- American College of Radiology Appropriateness Criteria: www.acr.org/secondarymainmenucategories/quality_safety/app_criteria.aspx.
- Berrington de Gonzalez A. Arch Intern Med. 2009; 169:2071-2077.
- Berrington de Gonzalez A. J Med Screen. 2008;15:153-158.
- Cedars-Sinai investigation. www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ucm185898.htm.
- National Lung Screening Trial Research. Radiology. 2010;doi:10.1148/radiol.10091808.
- Second Berrington de Gonzalez reference on Hemonc Today: www.hemonctoday.com/article.aspx?rid=51425.
- Smith-Bindman R. Arch Intern Med. 2009;169:2078-2086.
- Image Wisely. Radiation safety in adult medical imaging. Available at: http://www.imagewisely.org. Accessed December 8, 2010.
What are your concerns about the effect of radiation exposure in the use of CT scanning for cancer screening?
Risk from single CT scan is low
The experts are divided on the true risks of CT radiation exposure. There have been no studies showing that a CT scan has led to cancer in a patient. As there are many risk factors for developing cancer, including age, it gets very difficult to pinpoint a cancer to CT as a cause. Therefore, the estimations of potential cancer risk are only estimations and not based on hard data.
It is known that radiation can cause cellular damage that can lead to cancer. What is not known is the effect of cumulative low exposures due to medical imaging. The NEJM articles were based on data extrapolated from atomic bomb survivors, which many experts believed is a flawed comparison due to the type of exposure (atomic bomb does not equal CT radiation) and data analysis.
In general, the risk of a single CT scan is very low. In an emergency situation, such as after a car accident, questionable stroke or pulmonary embolus, the scan can be life-saving and should not be refused. Patients who get multiple CT scans in a nonemergent situation should discuss the necessity of the scan with their doctor, whether the scan can be performed with a low-dose technique and if the same question can be answered as effectively with a test that does not require radiation.
Multiple factors can lead to unnecessary radiation exposure including scanning errors from the radiology technologist or radiologist, unnecessary scans to improve reimbursement, or the practice of “defensive medicine” from medicolegal concerns resulting in ordering of low yield imaging exams
There has been appropriate attention brought to the fact that many practices need to consider dose when performing or ordering tests, and that manufacturers of CT need to supply tools that prevent overexposure errors.
Many radiology practices are looking at ways they can reduce radiation dose while preserving diagnostic quality that is best for the patient. Patients and clinicians are discussing if these scans are truly necessary, resulting in less defensive medicine being practiced. On the negative side, there is a lot of confusion on the patient’s side from the media attention, causing unnecessary fear from a test that can improve early diagnosis of significant disease.
Amy K. Hara, MD, is an associate professor of radiology at the Mayo Clinic in Arizona.
False positive results may lead to over-treatment
CT scanning has really improved the quality of medical diagnosis. It has reduced the need for testing, reduced incorrect diagnosis and improved the quality of medical care.
However, among other concerns is the frequency of false positive results that, in turn, result in over-diagnosis of disease. There is simply too much treatment. The medical community is enamored with medical imaging and focused more on the benefits than the harms. We need to consider this. We need to raise awareness of these risks.
There are two separate ways to think of radiation exposure. The first is radiation overdosing in an individual patient. This is usually the result of an error at some level, as in the case at Cedars-Sinai. In general, these kinds of errors happen in a small number of cases.
The other way to think about the radiation exposure is the widespread use of imaging. The more people who receive imaging, the higher the impact is going to be at a population level. When you couple increased dose per scan in the realm of too many patients, the risks are going to increase.
A final factor in this is that many people — particularly cancer patients — receive multiple scans over the course of a lifetime or even in the space of a few years. This contributes to both the individual risk and the population risk. It begins to add up, and the risk is no longer trivial on either level.
Regarding repeated imaging of cancer patients, what is happening is that while we are curing one cancer we may be putting them at risk for another one. The protocols for cancer surveillance are really aggressive right now.
The good news is that although we have seen wide variations in dose, and though there is evidence of gross overdiagnoses, the problems are beginning to be addressed. Media exposure, both positive and negative, has helped. The Image Wisely campaign will help. Patients are becoming more knowledgeable. It is important to give them all the information we have to help them make good decisions with their caregiver and radiologist.
We have acted for many years as if we do not need to worry about the radiation exposure. Early on, we leaned toward over-imaging, but now we are pulling back. As we move forward, I would like to see us focus more and more on alternatives, particularly prevention. Tell people to quit smoking or, better yet, never start. Place more focus on everything from colonoscopy to better dietary habits. Medical imaging is effective, but there are ways to tackle these problems without exposing patients to the risk.
Rebecca Smith-Bindman, MD, is a professor in residence in the departments of radiology, epidemiology/biostatistics, and obstetrics, gynecology, and reproductive medicine at the University of California San Francisco.