Prior to engaging in any discussion of imaging and radiation risk, the first point to emphasize is that computed tomographic (CT) scanning is a powerful and indispensable diagnostic tool. CT, more than any other imaging modality, has heralded a new age of medical practice based on the ability to visualize the inner workings of the human body.

As part of my teaching to residents, I tell them that, after 2 weeks of experience in CT scanning, a first- year radiology resident is a better diagnostician than any of the greatest clinicians in the pre-CT era, including Sir William Osler and everyone else; medical imaging is that powerful a tool, and CT has been the modality that has most represented this new diagnostic age.

The above being said, the indisputable fact, and in my opinion rendered truly indisputable by the BEIR VII report, is that medical x-rays cause cancer. BEIR VII also suggests that there may be no safe lower limit.

Imaging and Cancer: Kids Most in Peril

The implication that there may be no safe lower limit is cause for concern. In my opinion there is a common belief amongst radiologists that radiation exposure resulting in a dose below the dose of a standard body CT, which is approximately 10 mSieverts (mSv; units that radiation dose is measured in), poses negligible, if any risk, for an individual test.

However, even 1 body CT scan (1 CT scan of only 1 of the following regions: the chest region, the abdominal region, or the pelvic region) carries with it some element of risk. The risk that BEIR VII reports is 1 in 1000 chance of developing cancer from a 10 mSv radiation dose. In my prior report, I described what is written on the US Food and Drug Administration Web site, which is a 1 chance in 2000 of developing cancer from a dose of 10 mSv. The BEIR VII report doubles that risk. The risk in children is even higher, with a reported chance of 1 in 550 of developing cancer.

Risk escalates with multiphase acquisition in a single CT study, and with multiple individual CT tests. Because of their multiphase nature, careful limited use of multiphase renal CT studies and liver CT studies should be considered.

Another critical aspect of risk is this question: What kind of cancers (malignancies) are triggered by imaging radiation, and when do they occur? As preface to this, the answer in part explains why this information is not generally being conveyed, to date, to patients prior to their undergoing a CT study.

First, the cancers are relatively common in the general population, which means that it is difficult to sort out which cancers have occurred as a direct result from x-rays (unlike, for example, mesothelioma, which is an extremely rare cancer, and therefore its association with asbestos exposure was relatively easy to ascertain); second, the latency period between exposure to x-rays and development of cancer may be extremely long, up to 20 years, which means that the association may be difficult both to deduce and to prove. BEIR VII (and other sources) reported that the malignancies most associated with x-ray exposure are leukemia, thyroid, and breast cancers. The latency period between exposure and development of malignancy they report as 2-5 years for leukemias (and other blood line tumors) and 10-20 years for solid tumors (sarcoma, breast cancers).

My personal opinion is that many patients today who are receiving multiple CT scans may well be getting at least comparable doses to subjects that have now developed malignancies from x-ray radiation received in the 1930s and '40s. And, similar to those days when the doses were unknown, the dose that patients receive today over a course of years of multiple CT scans is also completely unknown.