| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||
Editorial |
In this issue we reproduce edited versions of the four contributions that were made to the debate on the motion "There is no radiation risk to health at low doses" at the UK Radiological Congress in Manchester in June (UKRC 2004).
The basis of the debate was that the Ionising Radiation (Medical Exposure) Regulations 2000 [1] and associated Guidance Notes [2] constitute wide-ranging legislation controlling the use of ionizing radiation in medicine. Primarily under the headings of Justification and Optimization this has led to the introduction of many new requirements in our procedures and new levels of record keeping. No one can question that this causes a lot of extra work. So what is the experimental basis for these requirements?
In attempting to reduce a complex problem to one simple statement, we might say "Application of the principles of Justification and Optimization to the very low doses usually encountered in diagnostic radiology is based primarily on the advice of the International Commission on Radiological Protection (ICRP) that there is no safe level of ionizing radiation – the so-called linear no threshold (LNT) model.
But how robust is the science underpinning this advice?
The problem is that nearly all the hard data on radiation-induced cancer in humans comes from effective doses above 250 mSv. How can we extrapolate over two orders of magnitude to annual background and typical diagnostic radiological doses of 1–2 mSv? ICRP thinking is that after a dose and dose-rate correction factor at higher doses, extrapolation should be linear to zero dose. However, this view is being increasingly challenged by evidence that there may be a threshold or even beneficial effects at low doses. For further detail on extrapolation models see for example Dendy and Dixon [3].
Thus the purpose of the motion was to examine critically currently accepted thinking on this controversial issue.
The motion was proposed by Ludwig Feinendegen under the title "Radiation Hormesis – fact or fiction" [4]. He started by observing that radiation-induced cancer has never been recorded at the doses used in diagnostic radiology and suggested there is now good radiobiological evidence that this is not simply a statistical problem. He pointed out that cells, tissues and organisms are complex adaptive systems and that adaptive protection is continuously being stimulated by reactive oxygen species generated by exogenous and endogenous toxins. From here he went on to develop the argument that the balance between damage and protection generally favours protection at low doses, i.e. hormesis.
Ken Chadwick opposed the motion with a lecture entitled "A mechanistic approach to justify the contention that radiation risk is linear with dose from zero dose up" [5]. He pointed out that induction of chromosome aberrations, somatic mutations and cell killing can all be described by a linear-quadratic dose–effect relationship. DNA double strand breakage is regarded as the critical lesion, resulting in a dose–effect curve that is normally linear-quadratic but linear at very low doses from zero dose upwards. On this model there is no threshold dose for radiation risk.
John Cameron spoke third for the motion. Using as his title "Moderate dose rate radiation is good for your health" [6], he presented epidemiological data, some of which is either unpublished or not readily accessible in the literature, to support the hypothesis that we need increased background radiation to improve our health.
Finally, Colin Martin opposed the motion. Taking as his title "The LNT model provides the best approach for practical implementation of radiation protection" [7], he considered both practical and political implications. He explored some of the difficulties of introducing any alternative model for protection and concluded that this is not the right time to abandon the LNT approach to implementation of radiation protection in healthcare.
Another successful innovation at UKRC 2004 was electronic interactive voting. This proved both enjoyable for the audience and informative. Each speaker had been invited to suggest one question based on the theme of their talk. The questions and voting are shown at the end of each lecture. An interesting finding was that, whereas almost 50% of the audience were prepared to question the underlying science of the LNT model, 70% were in favour of retaining it for practical implementation.
The substantive motion "That this house believes there is no radiation risk to health at low doses" was defeated (For the motion 30%; against the motion 54%; abstaining 16%). This was a clear but not overwhelming majority, and was a fairly strong signal that this important question should be kept under review. Although it is politically expedient to retain the LNT model, this should not deter scientific questioning of the model independently of political or practical considerations.
Undoubtedly the most encouraging voting response was that 77% of the audience said the debate had improved their understanding of the issues involved.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
