More on "Is Computed Tomography Safe?"

The December 7, 2007 entry to this blog was titled “Is Computed Tomography Safe?” As is often the case with such a difficult question, the answer is yes and no. No—there are risks associated with any exposure to ionizing radiation, so no procedure is entirely safe. Yes—in most cases the risks are small enough that the benefits outweigh the risks. In order to answer this question more precisely, a large scale study with nearly one million patients was conducted over three years. The conclusions were reported in the August 27 issue of the New England Journal of Medicine. The abstract of Fazel et al.’s paper “Exposure to Low-Dose Ionizing Radiation from Medical Imaging Procedures” (NEJM, Volume 361, Pages 849–857, 2009) is reproduced below.

Background: The growing use of imaging procedures in the United States has raised concerns about exposure to low-dose ionizing radiation in the general population.

Methods: We identified 952,420 nonelderly adults (between 18 and 64 years of age) in five health care markets across the United States between January 1, 2005, and December 31, 2007. Utilization data were used to estimate cumulative effective doses of radiation from imaging procedures and to calculate population-based rates of exposure, with annual effective doses defined as low (less than 3 mSv), high (greater than 20 to 50 mSv), or very high (greater than 50 mSv).

Results: During the study period, 655,613 enrollees (68.8%) underwent at least one imaging procedure associated with radiation exposure. The mean (±SD) cumulative effective dose from imaging procedures was 2.4±6.0 mSv per enrollee per year; however, a wide distribution was noted, with a median effective dose of 0.1 mSv per enrollee per year (interquartile range, 0.0 to 1.7). Overall, moderate effective doses of radiation were incurred in 193.8 enrollees per 1000 per year, whereas high and very high doses were incurred in 18.6 and 1.9 enrollees per 1000 per year, respectively. In general, cumulative effective doses of radiation from imaging procedures increased with advancing age and were higher in women than in men. Computed tomographic and nuclear imaging accounted for 75.4% of the cumulative effective dose, with 81.8% of the total administered in outpatient settings.

Conclusions: Imaging procedures are an important source of exposure to ionizing radiation in the United States and can result in high cumulative effective doses of radiation."

To help put this study in context, the NEJM published an accompanying editorial by Michael Lauer (“Elements of Danger—The Case of Medical Imaging,” Volume 361, Pages 841–843). Lauer writes that

Because the use of ionizing radiation carries “an element of danger in every . . . procedure,” we need to adopt a new paradigm for our approach to imaging. Instead of investing so many resources in performing so many procedures, we should take a step back and design and execute desperately needed large-scale, randomized trials to figure out which procedures yield net benefits. This approach would require leadership and courage on the part of our profession, our opinion leaders, and the research enterprise, but were we to insist that all, nearly all, procedures be studied in well-designed trials, we could answer many critical clinical questions within a short time. Because we will continue to be uncertain of the magnitude of harm, an accurate understanding of the magnitude of benefit is a moral imperative.

In Chapter 16 of the 4th edition of Intermediate Physics for Medicine and Biology, Russ Hobbie and I discuss the risk of radiation. While we do not provide a final answer regarding the safety of CT, we do outline many of the important issues one must examine in order to make an informed decision. The safety of computed tomography and other diagnostic imaging procedures will continue to be a crucial question of interest to readers of Intermediate Physics for Medicine and Biology. I will try to keep you posted as new information becomes available.

P.S. Thanks to Russ Hobbie for calling my attention to this paper. He reads the New England Journal of Medicine more than I do.