Medical-imaging procedures always worth the risk?

Medical-imaging procedures such as computed-tomography (CT) and myocardial-perfusion scans are up drastically from just 15 years ago, according to a new study published in The New England Journal of Medicine.

CT scans alone are up four-fold, according to the study. These "worrisome" radiation doses--as many as 2 percent of cancers could be attributed to radiation during CT scans alone--justify more rigorous scientific scrutiny, according to lead investigator Dr. Reza Fazel at the Emory University School of Medicine in Atlanta:

Unlike the exposure of workers in health care and the nuclear industry, which can be regulated, the exposure of patients cannot be restricted, largely because of the inherent difficulty in balancing the immediate clinical need for these procedures, which is frequently substantial, against the stochastic risks of cancer that would not be evident for years, if at all.

Almost 1 million adults between the ages of 18 and 64 in five U.S. cities took part in the study between 2005 and 2007, with 655,613 of the adults undergoing one or more imaging procedures associated with radiation exposure. This puts the mean effective dose at 2.4 millisievert (mSv) per person per year, but that includes a wide range of differentiation, often along gender, age, and geographic lines.

Myocardial perfusion imaging alone accounts for 22 percent of the radiation dose from all study procedures, and CT scans of the abdomen, pelvis, and chest make up for about 38 percent, according to the study. Women and the elderly are exposed to the highest levels; for instance, while only 50 percent of adults ages 18 to 34 underwent a procedure, 86 percent of adults ages 60 to 64 did.

While most adults received less than 3 mSv annually, "effective doses of moderate, high, and very high density were observed in a sizable minority," according to the study. (Just more than 19 percent of those enrolled were defined as having "moderate" exposure, with just less than 2 percent having "high" exposure.)

With few exceptions, like mammography, most radiologic testing offers net negative results because there is little high-level evidence that cumulative exposure can produce real harm, such as cancer caused by testing, according to Dr. Michael Lauer of the National Heart, Lung, and Blood Institute in Bethesda, Md., in an accompanying article, "Elements of Danger--The Case of Medical Imaging."

Lauer adds that these tests should be subjected to rigorous clinical trials and that physicians should offer radiologic imaging with the same "humility" as when offering experimental treatments, taking a careful history of the patient to determine the cumulative dose of radiation already received: "If we began a national conversation about the dangers of ionizing radiation, it might cause enough discomfort to stimulate demand for high-quality evidence our patients deserve."

If as many as 2 percent of cancers could be attributed to radiation during CT scans alone, as recent estimates suggest, I'm surprised that not enough discomfort has been caused already.