CT Scanning; risk versus benefit

Publisher Website
Google Scholar
Abstract
Several years ago, Shrimpton and colleagues at the National Radiological Protection Board [1] published the result of a survey of CT use in the UK, and came up with the result that while 4% of radiological procedures involved CT, this accounted for 40% of the collective effective dose to the public from diagnostic radiology. Now, Dr Mettler and colleagues (in this issue of the journal) have come up with the astonishing figure that in a large representative university hospital in the USA, 11% of diagnostic radiological procedures are CT examinations but their contribution to the effective dose from diagnostic radiology approaches 70% [2]. One reason for the difference may be the greater availability of CT in the United States. According to Bahador [3], who conducted a survey in 1995, there are 26.4 scanners per million population in the USA, while in Britain (where CT was invented!) the comparable figure is only 6.2. A more compelling reason may be that the NRPB survey was compiled nearly a decade ago, and the dramatic increase in the use of CT since has been propelled by the development of spiral or helical CT, the speed of which makes its use more practical in younger or less co-operative patients. This brings us to the second interesting statistic from the Mettler survey, namely that 11% of the CT procedures are in children, a higher proportion than previously indicated by, for example, the UNSCEAR report [4]. While the use of CT is on the increase for all age groups, it is paediatric CT examinations that are increasing at the fastest pace and so must elicit the greatest concern. There are two reasons for this: (a) For a given CT examination, the effective dose is significantly larger in small children than in adults [5]. (b) Lifetime cancer mortality risks per unit dose vary greatly with age; children less than 10 years of age are seven times more sensitive than a middle-aged adult, as calculated by the National Academy of Sciences BEIRV (Biological Effects of Ionizing Radiations) Committee of the ICRP (International Commission on Radiological Protection) [6, 7]. These considerations led Brenner et al [8] to estimate that the lifetime cancer mortality risk attributable to the radiation exposure from a single abdominal CT examination in a one year old child is approximately 1 in 550, which is an order of magnitude higher than for adults. These risks are balanced, of course, by undoubted (and overwhelming) benefits, but the point is that the risks are not negligible. The frequency of CT procedures in children and the significant doses involved need to be considered in the light of a case-controlled study from Canada which showed an excess of leukaemia in children (0-9 years) who had received two or more radiological procedures [9]. The adjusted odds ratio was 1.78 (1.21-2.63) rising to 2.26 (1.20-4.23) for girls alone, reflecting the gender difference in sensitivity to radiation-induced malignancies seen in children exposed by the A-bombs. The authors did not discriminate high dose from low dose procedures, except to comment that most were `bone x-rays', and yet the spontaneous leukaemia incidence was doubled in children receiving two or more diagnostic x-ray procedures. It would be fascinating to know what the odds ratio would be for two or more helical CT scans, rather than just two more radiological procedures of any kind, and to integrate this knowledge with the new information concerning the frequency of CT scans that comes for the Mettler survey.

This publication has 3 references indexed in Scilit: