JACC: What are the real dose risks with cardiac imaging?

With the growth of cardiac imaging, there has been heightened concern about the resulting increased exposure to ionizing radiation. Andrew J. Einstein, MD, PhD, of Columbia University Medical Center and New York-Presbyterian Hospital in New York City, outlined the currently available research on radiation exposure in cardiac imaging in the Feb. 7 issue of the Journal of the American College of Cardiology and concluded that while strong data specifically addressing the risks of cancer from cardiac imaging don’t exist, increased risks can be projected from studies involving similar levels of radiation exposure.

“In recent years, intensive efforts have been initiated to reduce the ionizing radiation associated with cardiac imaging,” wrote Einstein. “It is now routine for publications addressing cardiac imaging to report radiation doses, and several studies have estimated cancer risks from a variety of cardiac imaging procedures. Concern about potential deleterious effects from radiation, specifically cancers, abounds, in some cases even leading to avoidance of essential procedures.”

Einstein wrote that there are two reasons ionizing radiation has become a concern in cardiology: rapid growth in procedure volumes and high doses incurred by some procedures. There has been a threefold increase in nuclear SPECT myocardial perfusion imaging volume since the mid-1990s and hundreds of thousands of CT angiogram studies are now performed each year, according to Einstein.

With regard to dose, Einstein said it is not uncommon to see effective doses from cardiac imaging procedures up 50 mSv in select patient populations.

“It is easy for patients undergoing a single cardiac imaging examination to receive the amount of radiation equivalent to 1,000 chest x-rays, a lifetime of screening mammograms or many years' background radiation,” wrote Einstein. “If one takes dramatically increased volumes and multiplies them by high doses, the net result is a public health problem.”

How dangerous are these levels of radiation exposure? That question is hard to answer because no data relating ionizing radiation specifically from cardiac imaging to cancer risks exist, but Einstein pointed to a number of other epidemiological studies involving similar levels of radiation to help illuminate the subject.

Three major very low dose studies–involving Japanese atomic bomb survivors, nuclear industry workers and pediatric patients exposed to x-rays in utero–all demonstrated an increased cancer risk at doses commonly received by cardiac imaging patients.

Einstein also cited the Quebec Post-MI Study, a retrospective study of more than 80,000 patients, which found that radiation exposure from cardiac procedures averaged 5.3 mSv/patient-year, with some procedures resulting in an effective dose of more than 15 mSv. The study authors estimated that for every 10 mSv of radiation there was a 3 percent increase in cancer risk over a mean follow-up of five years.

Since no direct evidence of increased cancer risk from cardiac imaging exists, projections from risk models based on previous research are used to develop estimates. The U.S. National Academies has an expert committee on the Biological Effects of Ionizing Radiation (BEIR), and their latest report, BEIR VII, developed a linear no-threshold-assuming risk model based primarily on the study of Japanese atomic bomb survivors. Studies applying the BEIR VII methodology to screening with coronary artery calcium scoring found a five-year screening cycle would be associated with 42 cancers per 100,000 men screened and 62 per 100,000 women, according to Einstein.

Current risk models are not perfect, however, as many assumptions must be made to account for differences in baseline cancer rates between patient populations. “Application of BEIR models to populations with decreased life expectancy in comparison to the general U.S. population, as is the case for many populations of patients undergoing cardiac imaging and intervention, will result in overestimates of radiation-attributable cancer risk, and consequently estimates of the total number of cancers attributable to such procedures may also be overestimates,” wrote Einstein.

Despite the assumptions of the BEIR VII models, they are the best current tools to estimate cancer risks associated with cardiac imaging. Future epidemiological studies of patients exposed to medical radiation will further enhance understanding of radiation exposure risks.

“Nontrivial risks that have been described in some scenarios underscore the importance of justification of all studies involving ionizing radiation, a goal towards which appropriate use criteria and guidelines can serve as valuable tools,” wrote Einstein.