Over the past few years, menacing headlines have blared from newspapers hinting at the dangers from medical radiation, with titles like "Medical Radiation Exposure Of The U.S. Population Greatly Increased Since The Early 1980s" and "Too Much Radiation From Medical Imaging?"

While a genuine concern, the focus on the dangers of ionizing radiation from imaging threatens to obscure the more serious risks of surgery and other diagnostic procedures, researchers warn in the April issue of the American Journal of Roentgenology.

"Too much attention to radiation is likely to distract awareness away from the many – generally, greater -- non-radiation risks associated with most medical procedures," the authors write, led by Stephen Balter, a medical physicist with the department of radiology and medicine at Columbia University Medical Center in New York.
The researchers also note that models for calculating radiation risks or creating policy are often poorly predictive for a patient's specific risks, because of variations in how procedures are performed from clinic to clinic and differences in individual risk based on age, sex and innate sensitivity to radiation.

Yet despite the trouble of extrapolating real risks from these models, some researchers have relied on dosing estimates and so-called radiogenic risk models -- such as the linear no threshold model, which assumes that every increase in radiation raises cancer risks -- to predict horrifically widespread cancer deaths among the public at large.

[Read more about the controversy surrounding LNT]

"Unfortunately, collecting and combining nominal effective doses from radiation usage statistics and then using the LNT model to forecast radiogenic risk is increasingly common," the researchers write. "Assertions that CT results in '10,000 cancer deaths in the United States' are often derived from the LNT model without accounting for the age and health status of the irradiated patients."

To truly grasp the problem, the researchers argue, it's best to match up the risks from imaging with related risks from performing or not performing a procedure. As an example, they point to appendectomies and the CT scans used to diagnose them, and measures used to detect heart disease.

Appendectomies are quite safe, with a mortality of only around one out of 300 for open surgeries and one out of 1,400 for laparoscopic surgeries, according to research cited by the authors. Still, that tallies up to about 20 deaths in the United States every year from the surgeries alone. But using CT scans to diagnose appendicitis -- the scans have nearly 100 percent accuracy, the researchers said -- doctors could prevent around 13 percent of unnecessary operations: that is, surgeries where the organ is removed but the patient does not actually have appendicitis.

Based on these calculations, theoretically, CT scans could save about three lives every year. As for the risks from the ionizing radiation given off by the CT scans themselves, using conservative and highly theoretically estimates, the scans would be linked to two fatal cancers every year.

One (theoretical) life saved a year might seem like a small gain (if the life isn't yours, of course), but it comes on top of other benefits, such as better differential diagnosis and an average $450 per patient financial savings, the authors said.

And risk comparisons also apply to other diagnostic techniques, such as for heart disease. While coronary CT angiographies -- used to detect coronary artery disease -- deliver twice the effective radiation dose of conventional coronary angiographies (10 mSv vs. 5 mSv), they also eliminate the risks of bleeding, heart attack and infection associated with traditional angiographies, which involve snaking a catheter up an artery. "Coronary CTA may be a better choice when there is a clinical need to directly visualize the coronary arteries but a low expectation of finding treatable disease," the authors write.

More Industry Headlines