| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||
Correspondence |
I read with a great deal of interest the discussion on relative biological effectiveness (RBE) for mammography X-ray beam energies presented in the October edition of the BJR [1–4]. Such discussion demonstrates that the scientific basis of radiation protection of the patient is alive and well, and the BJR is playing a leading role in its stimulation.
Discussion on possible increased RBE for low energy X-ray beams based upon in vitro studies is unlikely to lead to a practical reappraisal of risk/benefits arising from the UK Breast Screening Programme (UKBSP) in the near future. However, it is right and proper that frank and open discussions on the basic principles and underlying methodologies employed in such studies are undertaken by the scientific community at an early stage. Whilst the studies that prompted this discussion were aimed primarily at possible effects on breast screening [5], the fundamental principles outlined apply equally to more general radiographic imaging procedures that involve the female breast.
In the first national survey of doses to patients undergoing a selection of routine radiographic examinations in the UK [6], the mean radiograph tube potential (kV) employed for the anteroposterior (AP) projection of the chest was 65 kV (range 56–91 kV), whilst for the posterior-anterior (PA) view it was 70 kV (range 42–123 kV). In a more recent review [7], the mean tube potential was 76 kV (range 60–95 kV) for the AP projection and 85 kV (range 50–150 kV) for the PA view. For X-ray beams generated in the range 65–80 kV, the mean beam energies are roughly 31–35 keV with a significant fraction of the spectral air kerma lying below 30 keV in all cases [8].
The energy carried by an X-ray beam per unit of exposure increases roughly by a factor of almost 10 for X-ray beams with mean beam energies in the range 31–35 keV compared to one of roughly 15 keV, applicable to that employed in mammography. This, of course gives rise to the increase in Dance conversion factors (g), used to convert incident air kerma to mean glandular dose for a standard breast phantom, with increasing HVL of the X-ray beam [9]. For example, for a standard breast thickness of 4.5 cm, this factor increases by a factor of 4.5 for an X-ray beam with HVL of 2.0 mm Al compared to one of 0.25 mm. However, the probability of cancer incidence is known to increase with imparted energy and for those types of examination involving a significant breast dose, the estimated cancer risk is higher for females than males [10]. This has already been noted for chest X-rays, barium meals and intravenous urograms (IVU) [10].
In 1977 it was noted that roughly 7 million chest radiograph examinations were undertaken in the UK [11] with almost 90% performed on female patients [12]. These were distributed equally across all age groups [13], including paediatric patients. The majority of these examinations would be performed in the PA projection, but probably around 10% would be performed in the AP projection. The female breast would, of course, be involved in all exposures. In the PA projection, roughly 10% of the surface entrance dose may be absorbed in the breast, whereas for the AP projection this will be over 50% [14]. Thus the RBE of low energy X-ray photons may also be of some relevance in more general X-ray imaging of female patients, including barium meal and IVU examinations.
Consequently, for an AP radiographic examination of the chest the total energy absorbed in the breast may be commensurate with that involved in a mammographic examination and a significant fraction for PA examinations due to the much higher g values. It is worth noting, however, that the energy fluence per unit of exposure increases with energy up to roughly 50 times for those employed in Volume CT compared to mammography beams. Scattered dose to the breast will also be of importance for these beam energies.
Irrespective of the relevance or otherwise of the RBE with beam energy, the relative sensitivity to irradiation of the female breast is significantly greater than that for male patients. Purely on the basis of relative breast masses, the sensitivity for the standard female patient, based upon data presented by ICRP [15], would be 11–15 times that for the breast of the standard male patient. However, the data presented by ICRP was based upon information derived prior to the 1960s. Given the known average increases in female breast size, which have been noted by the clothing industry in the intervening period, a much higher figure (>20) may be appropriate for a large group of females. However, It is unlikely that the relative sensitivities of breast tissue will only be dictated by relative breast size, but also by more inherent radiobiological considerations. Consequently there may well be sub-groups of female patients with significantly elevated risks from exposure to ionizing radiation during radiographic examinations involving the breast. Any possible increased RBE with beam energy would merely support the need to ensure that such groups are given due consideration in any radiation protection strategies.
The overall probability of a fatal somatic effect from chest radiograph examinations is possibly 2 times higher for female patients, and for non-fatal almost 3 times higher compared to male patients [13]. However, these data are not organ specific and have been averaged over all age groups, involving a so-called reduction factor [13]. Consequently, all of the concerns that were raised in respect of possible age extensions of breast screening programmes [5], I believe, apply equally to chest and other radiograph examinations of female patients, particularly young patients with larger breast sizes and dense breast tissue who may receive frequent examinations.
The situation for chest radiography outlined above was based upon the use of screen film systems. With the advent of digital imaging technologies, further considerations also arise. For example, it has already been shown that optimum beam energies for chest imaging using computed radiography (CR) systems for a standard sized chest (patient weight 70 kg) is in the range 75–90 kV [16], i.e. lower than that for screen film systems. For smaller (female) patients it will probably be lower still. Also of concern is the fact that digital imaging systems permit the use of standard exposure factors, dictated by larger (male) patients, irrespective of patient size.
Scope exists, therefore, for female patients to receive a relatively higher exposure than would ideally be required for a female population. In fact, this has already been noted in respect of body CT [17]. For volume CT studies of the thorax, the estimated overexposure of female patients compared to males was noted to be 67% and for the abdomen 87%. Obviously some form of specific control of exposures to female patients would seem desirable. At present, dose reference levels (DRLs) are dictated by the exposure of male patients and purely by considerations of body weight. However, for chest radiology, approximately only 10% of examinations appear to be undertaken on male patients for whom exposure of the breast is of minimal consideration. Perhaps DRLs should be set for particular examinations and specific groups of patients based upon the principle of overall equivalent risk?
Some may argue that risks from most diagnostic radiological procedures are so low that factors such as age, sex and specific organ exposure do not necessitate consideration. However, any approach that is adopted should be universal, rigorously scientific, logical and demonstrably aimed at ensuring the protection of those most at risk. At present this is patently not the case, even for high dose examinations such as barium meal and volume CT.
For a frequent examination such as the radiographic chest, the total annual UK effective dose is over 100 man Sv. Its actual magnitude will depend upon the value of tissue weighting factor that has been ascribed to the breast in published effective dose calculations (0.12 or 0.05) [10, 18] as well as the number of AP examinations. This is delivered predominantly to female patients of all ages and probably 25% of this exposure concerns risk to a single organ. This collective effective dose is over 25 times the total occupational exposure from all medical uses of ionizing radiations (40 000 staff) and over 4 times that for all the nuclear industry (41 000 staff) [19]. In the case of volume CT involving the chest, the collective effective dose may be orders of magnitude higher. However, the monetary investment per man Sv of collective dose reduction in the nuclear industry significantly exceeds that expended in medical radiology [20].
The "one size fits all" philosophy that at present appears to underpin radiation protection of the adult patient in medicine seems to be at odds with philosophical developments in other fields. For example, in the field of molecular imaging:
Molecular imaging techniques may also allow us to tailor therapy to individual patients. Visualization of specific molecular targets prior to therapy may identify patients most likely to benefit from a particular molecular therapy. [21]
The discussions presented in the October edition of the BJR, although scientifically relevant, are obviously somewhat esoteric in relation to the existing practicalities of radiation protection within the UKBSP. However, the points raised above are, I believe, of direct practical relevance and concern to over 50% of the patient population who undergo radiological procedures as well as the scientific community that supports these activities.
Yours sincerely
Director, Radiation Protection Service, Liverpool
Received for publication October 24, 2006. Accepted for publication November 20, 2006.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
