Fear of radiation from cardiac imaging outpaces knowledge about the risks

Reed Miller

January 25, 2011


New York, NY - Physicians and patients have little solid data on which to estimate the long-term risks of radiation from cardiac imaging, despite a recent surge in research and public attention on the link between imaging and cancer risks. As the number and range of cardiac imaging procedures continue to grow, so do fears about what all that radiation is doing to patients, even as some of the leading thinkers in this arena are coming up with increasingly creative ways to limit radiation exposure.

Dr Andrew Einstein [Source: Columbia University]

In 2009, the National Council on Radiation Protection and Measurements released a report showing that medical procedures had overtaken radon as the leading cause of radiation exposure for Americans. "That really woke us up," Dr Andrew Einstein (Columbia University, New York, NY), told heartwire . "A lot of us in the field knew that would someday be the case, but it woke up the cardiology community to the fact that . . . we needed to manage this burden better so as to optimize the care of our patients."

Einstein has led several studies of radiation exposure from cardiac imaging, including a 2007 study showing that the lifetime adjusted risk of cancer in a 20-year-old woman from a standard cardiac scan can be as high as one case for every 143 women[1]]. More recently, Einstein and colleagues reported that almost one-third of patients undergoing myocardial perfusion imaging at Columbia University Medical Center received a cumulative estimated effective dose from all medical sources of over 100 mSv, a level believed to be associated with an increased cancer risk[2].

Despite these kinds of numbers, Einstein believes physician knowledge about radiation is limited simply because it's not a significant part of medical school education or clinical training in cardiology. In the absence of that knowledge, people tend to overreact to radiation concerns or they ignore the problem completely, he says.

We needed to manage this burden better so as to optimize the care of our patients.

But studies like those by Einstein et al have begun to raise concerns about radiation risk among some patients referred for cardiac imaging, according to radiologist Dr Elliot Fishman (Johns Hopkins University, Baltimore, MD). "It's only been in the past year or so that patients or referring doctors would speak to you [about radiation]," Fishman says.

Dr Elliot Fishman [Source: Johns Hopkins University]


Right now, unfortunately, patients and physicians have to make decisions about exposing themselves to radiation based on only rough estimates of their risk, but there has been a lot of unjustified fear among patients stoked by media reports of these estimates, Fishman said. Many of the articles that scare patients about the cancer risk from cardiac imaging fail to point out that the patients being scanned are far more likely to die from the cardiac disease the scan is intended to diagnose than they are from cancer caused by the radiation, Einstein said.

"There's a lot more to be learned in this area. The field has not been as data driven as it should be, both in terms of observational data and data from randomized controlled trials," Einstein said. "Radiation is an important factor that needs to be considered in the design of future trials so that we can make decisions in a data-rich environment, which we're not operating in currently."

Einstein points out that the radiation from computed tomography (CT) is getting most of the attention and scrutiny because the number of coronary CT angiography scans performed in the US has grown so rapidly in recent years, but, he points out, a recent study by Dr Jersey Chen (Yale University, New Haven, CT) and colleagues showed that myocardial perfusion imaging (MPI) and diagnostic cardiac catheterization and PCI procedures account for the bulk of cardiac imaging radiation exposure. The volume of MPI increased from less than three million procedures in the US in 1990 to 9.3 million in 2002 and now accounts for more than 10% of the entire cumulative effective dose to the US population from all sources except radiotherapy.

Dr Jonathon Leipsic [Source: University of British Columbia]

"The ability to reduce dose using perfusion imaging is not quite as robust as it is for CT right now," Dr Jonathon Leipsic (St Paul's Hospital, Vancouver, BC) told heart wire . Leipsic and Dr Brett Heilbron (St Paul's Hospital, Vancouver, BC) are preparing to publish a single-center study of the effective radiation doses from a variety of cardiac imaging modalities, including CT and nuclear imaging. But, Leipsic stressed, "ionizing radiation, from whatever the source, is something that we need to be thoughtful and cautious about. So it's not a specific modality that we should be more or less concerned about. We should explore dose-reduction techniques across all of our imaging modalities that use radiation."

Still a lot unknown

Radiation exposure can damage DNA in cells, which leads to mutations that cause cancer. However, the cells are usually able to repair that damage, so radiation is generally a very weak carcinogen compared with all of the others that people may encounter daily. So the risks from the relatively low doses of radiation from single cardiac scans such as CT or a coronary angiography are not precisely quantifiable.

Dr Thomas Gerber [Source: Mayo Clinic]


"Dose is a very difficult construct," Dr Thomas Gerber (Mayo Clinic, Rochester, MN) told heart wire . So far, all of the studies of patients' cumulative radiation exposure from cardiac imaging are based on estimated averages. "It's not an actual measurement. It's an estimate based on assumptions about the radiation sensitivity of certain tissues."

The studies assume the same radiation exposure for every patient undergoing a certain kind of scan, even though radiation may interact with different bodies very differently. For example, a heavier person would absorb more of the radiation, Gerber said. "So determining dose alone is a difficult issue, and if you try to translate that into the risk of cancer, it's also difficult."

Dr Reza Fazel [Source: Emory University]

Dr Reza Fazel (Emory University, Atlanta, GA) agrees that even with a good estimate of the dose delivered by a particular type of imaging scan, estimating the risk of cancer from that level of radiation is difficult. In the absence of more conclusive data on the link between low doses of radiation and cancer, physicians are generally divided into two camps. One holds that the conservative "linear-no-threshold" model—that even low doses of radiation may be harmful and the risk increases linearly with dose—is the best assumption to make in the absence of better data. The other camp believes that the body can withstand and repair very low doses—below 50 or 100 mSv.

The best available estimates of the cancer risk associated with different levels of radiation were compiled by the National Academy of Science's Biological Effects of Ionizing Radiation (BEIR) VII Phase 2 report. But all of those estimates are based on studies of Japanese atomic-bomb survivors, high-altitude air crews, or workers in the nuclear industry. There are very little direct data from cardiac imaging patients.

It is very hard, from a scientific perspective, to base everything you're doing from an incident 65 years ago—that was a little bit different from what it means to get a CT scan.

Fishman is among those who suspect the atomic-bomb data overestimates the cancer risk from small doses of radiation. "It is very hard, from a scientific perspective, to base everything you're doing from an incident 65 years ago—that was a little bit different from what it means to get a CT scan—and there was all types of radiation, it wasn't monitored, and it was a very ill population," he told heart wire . "But if you look at the science, they all show that if you radiate the cell, the cell protects itself and then repairs itself. So to say it's just an additive or linear effect and the cells don't repair is not science."

Dr Manesh Patel [Source: Duke University]

Dr Manesh Patel (Duke Clinical Research Institute, Durham, NC) is confident that the linear-no-threshold model does not reflect reality. "Most cancer biologists will tell you that there is not a linear relationship with a lot of the things associated with cancer. There is some relationship with the dose, but then there are also some other things that might jump you up and down on the curve. . . . It's not a direct relationship for every patient, because the variables include age, your genetic background, and other predisposing factors for cancer."

And although 50 mSv is the generally accepted maximum safe annual exposed dose for radiation workers and hospital staff, there have been cases where people exposed to much lower doses have developed types of cancer most strongly associated with radiation effects, he said.

Most of the current guidelines on radiation exposure are based on cumulative doses over time, but it is also possible that there are some so-called "stochastic" effects, whereby a single exposure to radiation can cause immediate and irreversible cell damage that leads to cancer or other adverse reactions over time, Patel said. Fazel agrees that this is a critical unanswered question. "Intuitively, you'd think that a protracted lower level of radiation might be higher risk than a higher dose in a short period of time, but there are not a lot of data to scientifically support that."

Radiation and "episodes of care"

This question is particularly important for cardiac imaging patients, because they usually undergo multiple scans in a single "episode of care" or throughout the "care pathway" for a particular set of symptoms or diagnosis, Patel explained. Patel was the senior author on a study that showed that acute MI patients receive about 17.31 mSv of ionizing radiation per hospital admission.

According to Einstein: "We do not have really good radiation epidemiology of how much exposure every patient is getting for every test in the United States and the nature of the variability in such dosimetry. We have good first-order knowledge of the scope of the issue, but we really need more detailed knowledge to fine-tune the system so that benefits and risks can be balanced against each other."

Patel said the current estimates of patients' exposed dose are "very crude," but he "can see a day in the future where we put a dosimeter on them or those clip badges that we do in the cath lab so we can collect what they're actually being exposed to. That's a ways off, but I can envision that we're going to collect those types of data on our patients." He pointed out that some third-party payers are already pushing for collection of this kind of patient-specific information. And California recently passed a law that, as of 2012, will require radiation doses from CT exams to be recorded on the image and in a patient's health record and to be reported to the patients, physicians, and the California Department of Health, making it easier to keep tabs on cumulative radiation doses.

But even if more precise estimates of patients' exposure can be obtained, precisely defining the cancer risk associated with a given level of radiation will remain extremely difficult. A 2007 study by Dr David Brenner (Columbia University, New York, NY) and colleagues showed that the lifetime risk of dying of cancer caused by a single abdominal CT of 240 mA-seconds is about 0.14%[3]. About 40% of the population will eventually have some kind of cancer, and 20% of the population will die from it. So, Gerber estimates, it would take a prospective study of about five million people followed for their lifetime to definitively show a statistically significant difference in cancer deaths caused by CT scans.

"It would be logistically extremely difficult, and I wouldn't want to be the one to organize that study, but who is to say what is possible or impossible?" Gerber said. It is a question that could be followed by registries if radiation exposure could be routinely collected for each patient and recorded in their electronic medical record, he said.

CT is the leading concern

Most of the attention on radiation exposure in the past five years has been on cardiac CT angiography, since the number of CTs has been growing so rapidly, Einstein said.

This concern has led to a widespread effort by cardiologists, radiologists, and device manufacturers to lower the exposed dose from CT scans without reducing the image quality, with better equipment and new techniques to pinpoint the best moment to capture an image and reduce excess radiation exposure by reducing the duration of the scan.

Fishman said that almost all centers performing these tests are aware of the techniques for reducing radiation exposure. He acknowledges that equipment can be a barrier to dose reduction. "It's a lot easier to do this on a dual-source than a regular 64-slice scanner," he said. All of the vendors at recent radiology meetings, he noted, are pushing dose reduction as a major selling point for their equipment.

Einstein contends, however, that there are not a lot of data on how well centers are actually implementing dose-reduction methods. "There's not uniform availability of the latest technology that enables dose reduction or the best dose reduction, and there are still limited data from any of these dose-reduction methods, so there are a lot of the people who are not comfortable pushing things to their lowest limits," he said. He expects more operators to lower the exposed dose in the coming years as more centers buy newer equipment and become better educated on new techniques.

Education, quality, appropriateness

The experts interviewed by heart wire all expect the major professional societies, including the American Heart Association, American College of Cardiology, and the American College of Radiology to expand their role in managing radiation from cardiac imaging through new guidelines, accreditation standards, and appropriateness criteria. For example, the Society for Cardiovascular Angiography and Interventions (SCAI) recently published guiding principles and best practices for the development of radiation safety programs in cardiac catheterization laboratories[4]. The document includes guidelines on appropriate training and education, dose management and monitoring, equipment considerations, and potential safety concerns.

Each specialty professional group "will work with their constituency on more specific ways of reducing dose from the particular type of scans that they do," Gerber said. "Every level of professional society has a different role in bringing radiation-dose reduction to the attention of physicians at large and in the community." Fishman added that a lot of the excess radiation exposure from imaging could be reduced by simply reducing human error and through better quality control at the centers performing the tests. "You have to work hard to eliminate the chance of error," he said. "It's not just about the technology but about being able to use the technology well."

Leipsic said that clinicians often underestimate the exposed doses, so simply tracking each patient's dose exposure can do a lot to motivate physicians and staff to prioritize dose reduction. "If you don't know what the doses are, then you're making a number of assumptions and you run the risk of administering higher doses than is optimal for that clinical indication. There should be a mandate to 'know your doses,' " he said.

The most important radiation protection starts with the ordering of the test.

Although tracking and reducing radiation from imaging scans is important, all of the experts who spoke with heart wire emphasized that the most important question that every patient and doctor should ask before performing an imaging scan is whether the test is necessary.

"The most important radiation protection starts with the ordering of the test," Gerber said. Before conducting the test, physicians should ask themselves, "Do I need this information, and what am I going to do differently with a particular finding? Can I get this information with an imaging technique that doesn't employ radiation?"

"When a test is ordered, patients should be very comfortable asking why it is being ordered and how it is going to help manage their disease," Fazel said. He also suggested that patients find out if the center conducting the test is accredited. "Most of us like it when our patients are trying to understand what is going on and why things are being done. I think that's something that ideally every patient should ask."

"If the test is going to add significant value, then the risk/benefit analysis will be in favor of taking that potential risk, but on the other hand, if it doesn't meet the appropriateness guidelines, you shouldn't do it," Leipsic said.

Einstein reports serving as a consultant to GE Healthcare and receiving travel funding from Philips Medical Systems. Patel is an advisor to Genzyme. Leipsic has previously disclosed serving on the speaker ' s bureau for GE and Edwards Lifesciences. The other experts cited reported they had no relevant potential conflicts.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.