Various recently published studies about the relationship between low dose radiation and cancer fanned again the discussion about risk and benefits of modern diagnostic techniques and in particular Computer Tomography (CT). According to an article recently published in IMAGE (Marie H. Meynadier, Vol. 21, no. 10 - March 10, 2008), it is predictable that in a few decades up to 2 percent of cancers in the United States could be linked to the administration of diagnostic x-rays, even if the ability to monitor the increase in cancer related to radiation exposure will be very difficult as radiation-induced cancer can take up to 20 years to develop.

Since many years the relationship between cancer development and radiation exposure has been studied and there are hundreds if not thousands of publications related to this topic. The problem is that a third of all people get cancer anyway, at some time in their lives, and hence it is quite difficult to find evidence that low doses of radiation cause cancers that would not have otherwise occurred. Even for the 80,000 to 90,000 survivors of the atomic bombs exploded over Hiroshima and Nagasaki, exposed to very large radiation doses, it has been hard to find a direct relationship between excess cancer development and radiation exposure. According to Japanese statistics, from the people who were exposed in 1945 (and did not die immediately) nearly half are still alive. A statistically significant increase in cancer was found at relatively high exposure level of 50 millisieverts (mSv is the unit commonly used to measure the effective dose in diagnostic medical procedures), which is about 16 times the current annual average for Americans from medical exams and about 21 times above the average natural background effective dose in the USA. But these numbers need to be analyzed with care. There is a controversy between scientists if levels below 50 mSv can be considered as safe and about the real (and statistically proven) rest-risk of low radiation exposure.

The most widely used mathematical model in estimating radiation risk is known as the linear-nonthreshold dose-respond model. This model assumes that there is no safe dose of radiation and that there is a linear and direct relation between cancer risk and genetic damage with radiation exposure. But this model is in discussion since a long time. For some scientists the linear model is the best way to estimate radiation risk, but for others there exist a threshold below which radiation poses no hazard to health.

It is not difficult to identify the most important sources of man-made or anthropogenic radiation. Most of the collective dose from diagnostic radiology comes from procedures such as CT, interventional radiology and barium enemas. Advances in radiological diagnostic technology have radically transformed medical practice in recent years and there has been a rapid increase particularly in the application of CT. Effective dose estimates of CT scans and nuclear medicine studies are in the range of 10 to 25 mSv for a single study, which is about 100 times larger than those from conventional radiological procedures such as chest x-rays. Although CT scanners contribute to only 12 percent of all medical radiation procedures, the number of CT scans is on the rise. In the United States for instance, no more than 3 million studies using CT were performed in 1980 and in 2006 this number was already above 60 million. Overall, the mean effective dose in the US from all medical x-rays has increased about seven-fold over this period and the situation in Europe is not very different. There are several scientific studies proving a relationship between small radiation doses and cancer formation.

One of these studies, already published in the year 2000 (Spine, Morin Doody et al, Volume 25(16) August 15, 2000), was conducted with 5,573 females aged 20 and younger from 14 orthopedic medical centers in the United States who had been diagnosed with scoliosis between 1912 and 1965. Scoliosis is a medical condition in which a person's spine is curved from side to side, and may also be rotated. The goal of the study was to evaluate breast cancer mortality patterns among women with scoliosis and the risks associated with diagnostic radiographic exposure. Doody and coworkers showed that exposure to multiple diagnostic radiographic examinations during childhood and adolescence may increase the risk of breast cancer among women with scoliosis. Even if this study addresses radiology exams taken on older technology, the issue is still relevant, as scoliosis patients usually receive repeated radiographic exams and CT scans for therapy and treatment. As scoliosis generally is diagnosed even before adolescence, these susceptible young patients are already exposed to relatively high radiation doses during their growth period.

As discussed in an other recently published study (Hall et al., British Journal of Radiology 81, 362-378, 2008), the relevant organ doses during CT scans are in the range for which there is now direct credible epidemiological evidence of an excess risk of cancer, without the need to extrapolate risks from higher doses. However, even for high-dose radiological procedures, the risk to the individual patient is small, so that the benefit/risk balance is generally in the patients' favour. Concerns arise when CT examinations are used without a proven clinical rationale, when alternative modalities could be used with equal efficacy, or when CT scans are repeated unnecessarily. According to this study, it is assumed that about one-third of all CT performed in the US could be avoided using alternative diagnostic tools. According to Hall, “at this time, the benefit/risk balance for any of the commonly suggested CT screening techniques has yet to be established.”