CT Scans May Deliver Higher-than-Expected Radiation Doses
Published: December 14, 2009
* Explain to interested patients that the use of CT scans has increased dramatically from about 3 million in 1980 to about 70 million in 2007.
* Point out that the radiation dose delivered from a CT scan is much higher than that delivered by conventional radiography.
Radiation doses from computed tomographic (CT) scans are higher than previously thought and vary widely, even among the same types of examinations, two studies showed.
In the first study, based on information from four San Francisco-area hospitals, median effective doses ranged from 2 mSv for a routine head scan to 31 mSv for a multiphase abdomen and pelvis scan, according to Rebecca Smith-Bindman, MD, of the University of California San Francisco, and colleagues.
Radiation doses as low as 10 mSv have been linked to an increased cancer risk among survivors of the Hiroshima and Nagasaki atomic bomb blasts, the researchers reported in the Dec. 14/28 Archives of Internal Medicine.
The issue is important, investigators say, because over the past two decades, the use of diagnostic CT has skyrocketed: from 3 million scans nationwide in 1980 to 70 million in 2007.
In the latest study, exposure varied widely both within and between hospitals, with a median 13-fold difference between the highest and lowest dose for each study type, "highlighting the need for greater standardization across institutions," they wrote.
In addition to standardizing procedures, they suggested reducing the number of unnecessary CT scans and tracking dose information for individual patients to lower overall radiation exposure from medical imaging.
A second study, by Amy Berrington de Gonzalez, DPhil, of the National Cancer Institute in Bethesda, Md., and colleagues, projected that 29,000 future cancers will be directly attributable to CT scans performed in 2007.
In an accompanying editorial, Rita Redberg, MD, of the University of California San Francisco, editor of Archives of Internal Medicine, said the two studies "make us question if we have gotten carried away in our enthusiasm" for the use of CT.
It's becoming clear, she said, that the large doses of radiation from CT scans will lead to additional cancers, which must be taken into account when physicians consider use of CT for their patients.
"In light of these data, physicians (and their patients) cannot be complacent about the hazards of radiation or we risk creating a public health time bomb," she wrote.
In a statement, Robert Smith, PhD, director of cancer screening for the American Cancer Society, urged caution when interpreting the results of the studies.
"As striking as these numbers are, we need to address this issue with common sense," he said. "It is important that patients not refuse a CT scan when one is indicated."
But he said the results have to be taken seriously.
"They remind us that the potential benefits of a CT scan have to outweigh the possibilities of harm."
To quantify the doses and associated potential cancer risk from various types of examinations, Smith-Bindman and colleagues conducted a retrospective analysis of the 11 most common types of diagnostic CT performed on 1,119 patients.
The median effective radiation doses associated with various exam types ranged from 2 mSv for a routine head scan to 31 mSv for a multiphase abdomen and pelvis CT scan. For coronary angiography, it was 22 mSv.
For a single multiphase abdomen and pelvis CT scan, the median effective radiation dose delivered was equal to about 74 mammography series and 442 conventional chest radiography series.
For a CT coronary angiogram, the dose delivered to the breast equaled about 15 mammography series and the dose delivered to the lung equaled about 711 chest radiography series.
There was wide variation in the median effective doses, even within the same type of scan, the researchers wrote.
"While some of this variation may be clinically indicated to accommodate patients of different size or the specifics of the clinical question that was being addressed, the variation in effective dose was dramatic and of greater magnitude than widely considered acceptable," they wrote.
The projected impact on future cancer risk varied by exam type, age, and sex.
For example, an estimated 1 in 270 women who undergo CT coronary angiography at age 40 will eventually develop cancer directly related to that scan; for men, the estimated rate was 1 in 595.
Risks were roughly doubled in 20-year-olds and halved in 60-year-olds.
"CT is generally considered to have a very favorable risk-to-benefit profile among symptomatic patients," Smith-Bindman and her colleagues wrote.
"However, the threshold for using CT has declined so that it is no longer used only in very sick patients, but also in those with mild, self-limited illness who are otherwise healthy. In these patients, the value of CT needs to be balanced against this small but real risk of carcinogenesis resulting from its use."
In the study by Berrington de Gonzalez and colleagues, the risk of future cancers from CT scans in 2007 was estimated using risk models based on the National Research Council's "Biological Effects of Ionizing Radiation" report.
Of the 29,000 projected cancers -- or about 2% of all cancer diagnoses -- most were expected to be caused by scans of the abdomen and pelvis (14,000), chest (4,100), and head (4,000). About 2,700 would be caused by CT coronary angiography.
The most commonly caused cancers would be those of the lung (6,200) and colon (3,500), as well as leukemia (2,800).
The projected number of cancers decreased with older age at exposure.
Acknowledging the estimates involved several assumptions, the researchers wrote that "further work is needed to investigate the balance of the risks and benefits from CT scan use and to assess the potential for dose or exposure reduction."
Smith-Bindman and her colleagues listed some limitations on their radiation dose estimates:
* The cohort was not large enough to explore the reasons for dose variations within each type of study.
* There was no assessment of the association between image quality and radiation dose.
* All scanners were from a single manufacturer.
* There was possible misclassification of the indications for each exam type
* Methods for determining effective dose were imprecise.
In an interview, James Thrall, MD, raidiologist-in-chief at Massachusetts General Hospital and chair of the American College of Radiology's board of chancellors, raised methodological issues involving the Berrington de Gonzalez study.
First, the projected cancer risks were based on a linear model in which increasing radiation dose, even at low doses, is associated with rising cancer risk.
He said there's no evidence that this model is superior to one in which there is little cancer risk until a certain threshold is reached.
Secondly, he said, is that the projections were made using the assumption that the patients undergoing a CT scan had the same life expectancy as the general population, even though most CT scans are for a medical problem.
The projections are likely overestimates, he cautioned, which could scare patients and physicians away from using necessary CT scans.
"Radiation should never be given unnecessarily," said Thrall, who noted that he, too, believes there are too many CT scans performed each year.
"If a scan is necessary, it should be done with the lowest dose possible," he said, "and before a scan is done the reasons for doing it should be challenged and subject to appropriateness review."
He also tried to put the 29,000 projected cancers into context. Because about a quarter of the population will develop cancer with or without radiation exposure, about 15 million people in the analysis would develop cancer anyway.
"In that context, 29,000 is a very small number when you put it up against the immediate benefits to the patients from the scans they receive."