| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Full Paper |
1 National Co-ordinating Centre for the Physics of Mammography, Medical Physics Department, Royal Surrey County Hospital, Guildford GU2 7XX and 2 Breast Test Wales, 18 Cathedral Road, Cardiff CF11 9LH, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionConclusionsReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
Measurements of doses to samples of about 50 women attending for screening are recorded routinely at most screening centres and a pilot study collected dose data from screening centres covering the years 1994 and 1995 [6]. A more recent paper used similar methodology to review the doses from screening centres in 1997 and 1998 [7]. This paper reviews new data collected on samples across the NHSBSP during 2001 and 2002 using a similar methodology but with the additional factors in Dance et al [5]. Where dose data are shown (or referred to for comparison) for 1997/8 they have been recalculated using the additional factors described in Dance et al [5]. The aims of this study were:
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
The survey was restricted to the basic screening situation and so more complicated procedures such as magnification mammography were excluded. Data on doses for women attending screening below the age of 50 years as part of a research trial are being collected and have been reported separately [8].
The exposure factors and compressed breast thickness for each mammogram in the sample were recorded by radiographers and supplied to their local medical physics services. These data were entered into a database supplied by the NHSBSP [9]. The physics service added further details including X-ray tube output data and the MGD per mammogram was calculated automatically. All the information was transferred to a central database for subsequent analysis. The view information was coded to distinguish between left and right breasts, mediolateral oblique projection (OB) and craniocaudal projection (CC). In addition, films were categorised as "main" films or "extra" films. The extra films arose where a breast could not be completely covered with a single film. Unless otherwise indicated, the first film was assumed to be the main film. Examinations were categorised as either 1-view (1V) or 2-view (2V). A 2V examination is conducted at the first visit in the NHSBSP and comprises OB and CC films. Until recently the NHSBSP recommendation was to conduct a 1V examination comprising OB films for all subsequent visits. The NHSBSP (in England) has undergone a phased introduction of 2V examinations for all screens by 2003.
The total dose for a screening procedure or "examination" was calculated by summing the doses for all films and averaging over both breasts. This involved summing the doses for both views in a 2V examination, but also adding the doses for any extra films that were taken to cover large breasts.
Where means for MGD or breast thickness have been calculated they are shown in the tables and figures with 95% confidence limits. The distribution of breast thickness was well represented by a normal distribution. As found previously the distribution of MGD was found to be well represented by a lognormal distribution. In practice it was found that the confidence limits for means of MGD were effectively the same whether a normal or a lognormal distribution was assumed. Thus the 95% confidence limits have been reliably estimated by using ± 2 standard errors of the mean.
The Ionising Radiation (Medical Exposure) Regulations 2000 require employers to establish DRLs for radiodiagnostic examinations [10]. DRLs are defined in the regulations as dose levels for typical medical X-ray examinations for groups of standard-sized patients or standard phantoms and for broadly defined types of equipment. The employer is required to establish DRLs and to ensure that procedures are in place for using them on the understanding that they are not expected to be exceeded for standard procedures when good and normal practice is being followed. The employer is, moreover, obliged to undertake appropriate reviews whenever DRLs are consistently exceeded and to ensure that corrective action is taken when the excessive exposures cannot be clinically justified. The Institute of Physics and Engineering in Medicine (IPEM) along with the National Radiological Protection Board (NRPB), the College of Radiographers (CoR), the Royal College of Radiologists (RCR) and the British Institute of Radiology (BIR) have established a Working Party to provide guidance on the implementation of DRLs for diagnostic X-ray examinations. The Working Party has proposed a dose audit measure for mammography which is the average MGD for mediolateral oblique mammograms for breasts with a compressed thickness of 55 ± 5 mm [11]. A minimum of 10 women should be included in the dose sample. The proposed National DRL (NDRL) for this dose audit measure is 3.5 mGy. In this study the average MGD for all oblique mammograms (main films only) with a compressed thickness of 55±5 mm was calculated for each dose survey. For a few surveys the number of OB mammograms in this thickness range was less than 20 (i.e. equivalent to less than 10 women) and these were excluded from this analysis.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
X-ray technique and systems used
Only 8% of the X-ray sets were used in a mode that selected the beam quality manually. The rest had some form of automatic selection. The tube voltage selected ranged from 23 kV to 34 kV. Where the tube voltage was selected manually a molybdenum target material was always used, and 97% of films were taken using a molybdenum filter and 3% using a rhodium filter. The average tube voltage selected in manual mode was 28.2 kV. The proportions of films taken with different target/filter combinations are shown in Table 1
and compared with those in 1997/8. All units operated with an anti-scatter grid, which is standard practice within the NHSBSP. Standard optical densities (including base and fog) ranged from 1.11 to 2.11 and averaged 1.71. 89% of the systems had densities that lay within the range 1.50 to 1.90.
|
. The average MGD for OB films decreased from 2.36 mGy to 2.23 mGy, but that for CC films increased from 1.86 mGy to 1.96 mGy. As before the MGD and breast thickness were lower for CC films. Mean thickness increased from 54.3 mm to 56.8 mm for OB and from 51.5 mm to 54.1 mm for CC films. The histograms of doses for OB films are compared in 0.2 mGy bands for 1997/8 and 2001/2 in Figure 1. In 1997/8 5% of films had doses in excess of 5 mGy. In 2001/2 this proportion had fallen to 2%. The average doses for 1V and 2V examinations are compared with the previous data in Table 3. The histograms of the doses for 1V and 2V examinations are shown in Figure 2.
|
|
|
|
. The MGD to the standard breast for the systems that provided age information was 4.2% higher than the average for all the systems and as a result the average doses in Table 4 are higher (by 4.5%) than in Table 2
. To minimize the effect of any confounding due to variations in the MGD to the standard breast for the different mammography systems, the average ratio of the dose for the main OB films to the dose to the standard breast was calculated for the three 5 year age bands and is shown in Table 4. This ratio declined by 4% between the age groups 50–54 years and 60–64 years (p<0.001, Student's t-test).
|
. The effect of thickness on MGD where automatic and manual beam quality selection was used is shown in Figure 4. Where manual beam quality selection was employed the MGD increased more rapidly with breast thickness. Average doses per OB film were found to be about 10% lower where X-ray sets were used with automatic rather than manual beam quality selection as shown in Table 5. However, the compressed breast thicknesses are also slightly lower. The average doses per CC film were about the same where X-ray sets were used with automatic rather than manual beam quality selection and the average compressed breast thicknesses slightly lower. The variation in dose with breast thickness for different models of X-ray set is shown in Figure 5. The highest doses were found with the Lorad Mark IV X-ray set (Hologic Inc., Bedford, MA). During the period of the data collection Lorad introduced a new version of their AEC software which was designed to reduce doses. The use of this software is optional and some users may choose to continue with the older version. The data in Figure 5 are shown separately for Lorad systems that use the old and new versions of the software.
|
|
|
|
shows the number of films per view per breast for those centres with and without the large format film. The average doses per examination for units with and without the facilities for large format film are compared in Table 7.
|
|
Diagnostic reference level
The average MGD for main OB films for 50–60 mm thick breasts was 2.03±0.08 mGy, with a minimum of 0.85 mGy and a maximum of 4.06 mGy, with a distribution as shown in Figure 6. Nine systems (3.5%) out of the 259 for which data were available exceeded the NDRL of 3.5 mGy. Seven of these systems were Lorad Mark IVs using the old automatic exposure control (AEC) software while the other two were another Lorad Mark IV using the new software, and a GE 800T. Figure 7 shows the MGD for 50–60 mm thick breasts against the MGD for the standard breast at 28 kV and a Mo/Mo target/ filter combination. The Lorad Mark IV X-ray sets are highlighted because they tended to have the highest doses. The regression line shown has a correlation coefficient of 0.54 and a gradient of 1.42. (The regression was set to intercept the origin and has a significance of p<0.001.) The systems that exceeded the NDRL were not generally the same systems as those that exceeded the 2 mGy limit to the dose to the standard breast.
|
|
. From these data a relative speed for each combination has been calculated. The film–screen combination used at a screening centre is one of the main determinants of the average MGD per film as shown in Figure 8. This figure also shows that if one multiplies the average MGD to the standard breast for a given film–screen combination by 1.57 the result approximates the average MGD for OB films. The figure of 1.57 is the average MGD for all main OB films, 2.23 mGy, divided by the average MGD to the standard breast at a tube voltage of 28 kV and a Mo/Mo target/filter combination, 1.42 mGy, for all film–screen combinations.
|
|
. The average doses to the standard breast for each model are also shown for a tube voltage of 28 kV and a Mo/Mo target/filter combination. The highest doses were for the Lorad Mark IV X-ray set used with the old AEC software. The new AEC software resulted in doses that were about 16% lower than with the old software, after allowing for differences due to variations in the dose to the standard breast.
|
. A sub-group comprising 0.8% of women screened had a compressed breast thickness greater than or equal to 90 mm for the OB view. A smaller proportion had breasts with a compressed breast thickness greater than or equal to 90 mm in the CC view. The average doses for these women are shown for both automatic and manual beam quality selections. The small proportion (about 9%) of these women who were screened in manual mode had average doses of 6.64 mGy per OB view, which is about 2.7 times the average dose of 2.47 mGy for all women screened in manual mode (Table 5). Those women screened in automatic mode had average OB doses of 4.84 mGy, which is about 2.2 times the average dose of 2.21 mGy for all women screened in automatic mode.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
Data included
The data analysed here are broadly representative of the NHSBSP and represent the largest study of mammographic doses ever published. The women were almost all in the normal screening age range 50–64 years with an average age of 56 years. About 76% of the NHSBSP programmes participated and data were collected for over half the X-ray sets used by the NHSBSP. The equipment included a wide range of X-ray models and film–screen systems. A few of the datasets submitted were smaller than 50 women. This did not cause a problem for most of the analyses since the data for many different systems were usually averaged. However, the comparison with the NDRL was not possible in a few cases where less than 20 OB films were included for 50–60 mm thick breasts.
X-ray technique
Modern machines with a choice of filters have replaced most of the X-ray sets used at the time of the dose review of 1997/8 and many have a choice of target material. Some of the changes in X-ray sets have tended to reduce doses (e.g. availability of rhodium filters) while others have increased them (e.g. selection of lower kV and grid designs that improve contrast but increase dose). In 1997/98 only 15% of the X-ray sets selected the beam quality automatically. In this review 92% of the sets were used in an automatic mode. One consequence of this change is the much greater use of a rhodium filter. Thus the proportion of films taken using a rhodium filter has increased from 3% to 40%. However although many modern machines have an alternative target material such as tungsten or rhodium, these are still relatively rarely used. This is because current guidance in the NHSBSP is that the use of these target materials with film–screen imaging can result in significant loss in contrast as well as a large dose reduction. Standard film densities have increased only slightly from 1.64 to 1.71 since the previous review.
Average doses
The average MGD per OB film has fallen by about 5% since 1997/8. The average MGD for CC films has increased by about 5%. The average thickness for both OB and CC films has increased by about 2.5 mm. This could be due to a real change in the sampled thickness or due to a change in the calibration of thickness displays. It is surprising that average doses have not fallen by more given the widespread use of rhodium filters. However there have been many other changes in the X-ray sets, film processors, films and screens used in the NHSBSP, some of which tend to increase doses. Doses for CC films were about 12% lower than for OB films. Previously CC films had doses that were about 19% lower than OB films. The small difference found in the compressed breast thickness of about 3 mm explains most of this dose difference. The other explanation of the difference is likely to be the effect of pectoral muscle in the oblique views overlying the AEC chamber and causing an increase in the exposure. Changes in AEC design may explain why the difference between the doses for OB and CC films has varied.
The dose for a one view examination was slightly higher than the dose per film because of the additional films per view shown in Table 6. The dose for a two view examination is also much higher because of the additional CC view. The change to two view examinations in every screening round causes by far the greatest change in doses to women attending for screening. Thus the MGD for a full screening examination is now about 4.3 mGy. The decision to change to two view imaging at every screening visit was based on the prediction of an increase in cancer detection [12]. A review of the radiation risks and benefits of such a screening regimen concluded that the benefits greatly exceeded the risks [13]. The dose assumed in that review was an average of 4.5 mGy for a two view screening episode which is consistent with the results reported here.
The average doses found in the UK NHSBSP can be compared with a few similar but smaller studies for other countries. Jamal et al calculated the MGD per film in Malaysia for 300 women examined using three X-ray units [14]. The average MGD was 1.54 mGy and 1.82 mGy for CC and OB films, respectively, which is lower than presented here. However the corresponding average compressed breast thicknesses were also much lower at 37 mm and 44 mm, respectively. Kruger and Schueler reported on patient doses for 6006 women undergoing mammography in the USA using 7 Lorad Mark-IV X-ray sets [15]. They calculated an average MGD per film of 2.6 mGy and a compressed thickness of 5.1 cm for all views combined. Such higher doses can be explained by the use of this particular X-ray set. The use of different conversion factors can also be expected to cause a difference. Bulling and Nicoll have reported on mammographic doses for a screening programme in New Zealand [16]. They reported on doses for 310 women, routinely screened using a GE Senographe 600 operated mainly at 25 kV with Mo/Mo target filter combination. The average MGD per film was about 2.8 mGy with an average thickness of 4.9 cm.
Age
Age information was available for 60% of the women, which allowed the relationship between age and radiation dose to be assessed. Since the system to system variations in dose were quite large it was important to ensure that these differences did not mask or exaggerate any relationship between age and dose. For each X-ray set information on age was usually provided for all the women or for none. The distribution of ages sampled for the different systems was not perfectly uniform, and it was therefore prudent to take the step of normalizing the dose data by dividing by the standard breast dose as in the last column of Table 4. Since it is known that breasts become less glandular with age, one might have expected dose to decline with increasing age. In the previous review a small reduction of about 3% in the normalized dose data between the age bands 50–54 years and 60–64 years was found but was not statistically significant. In this study a 3% reduction is again observed, and is in this case significant because of the larger sample size. A slight increase (3%) in breast thickness over this age range was observed as found previously. In a previous report no effect of age on doses was found when the doses for women in the age range 40–48 years were compared with those in the screening programme [8]. Overall, age appears to have only a very small effect on dose.
Compressed breast thickness and beam quality selection
The compressed breast thickness is a major determinant of the dose per film received by a woman for a given mammography system. Compressed thicknesses were recorded, which ranged from 10 mm to 100 mm with an average thickness of about 57 mm for OB films and 54 mm for CC films. These thicknesses are about 2.5 mm greater than reported in 1997/8 (Table 2). Possible reasons are a real increase in breast size due to increasing obesity in the population, less compression force being applied, or problems with calibration especially with new tilting paddle designs. In the authors' opinion the latter is the most likely.
The difference in thickness largely explains why doses are higher for OB than CC films. However Figure 3 shows that there appears to be an additional factor. For thickness greater than about 60 mm the OB doses were larger even for the same breast thickness. It is likely that the presence of pectoral muscle overlying the AEC chamber is the cause of greater doses for OB films than CC films of the same compressed breast thickness. In general there was a curvi-linear relationship between the compressed breast thickness and the dose per film. Where beam quality was manually selected (and usually fixed at a tube voltage of 28 kV with a Mo/Mo target/filter combination) the average dose per film rose from just above 0.9 mGy for a 20 mm thick compressed breast to 5.7 mGy for a 90 mm thick compressed breast. Where X-ray sets were used in an automatic mode the selection of a different tube voltage, target and filter reduced doses for breasts with a compressed thickness greater than about 60 mm. Thus in automatic mode large breasts with a 90 mm thickness received doses which were about 2.2 times that for average sized breasts (55 mm thick). However Figure 5 showed that the change in dose with thickness varied from one model of X-ray set to another. Lorad Mark IV X-ray sets programmed with the old AEC software typically selected 25 kV and a Mo/Mo target filter combination for average sized breasts. As a result the average dose for a 55 mm thick breast was 3.2 mGy. The new AEC software on the Lorad X-ray sets tended to select 28 kV and a Mo/Mo target filter combination for average sized breasts. As a result the average dose for a 55 mm thick breast was 2.7 mGy. For the women with relatively thick breasts on compression, automatic beam quality selection leads to a significant dose reduction. However, the average dose per film for all women screened using automatic beam quality selection shown in Table 5 was not much lower than that of women screened on systems using manual beam quality selection. This is partly explained by higher doses for women with thinner breasts and partly by the increase in dose caused by the use of Lorad X-ray sets. The interpretation of Table 5 is also complicated by the fact that the average compressed breast thicknesses were slightly lower where automatic systems were used. The small differences found may be real leading one to expect lower doses due to thinner breasts or just reflect thickness scale calibration differences between different X-ray sets.
Film size
Many screening centres in the UK do not have the facilities for the larger format 24 cm x 30 cm mammography film. Some women's breasts are too large to fit on the standard sized film and have to be imaged using a mosaic of the standard sized films. This procedure involves a higher dose than if a single large format film were used. The extra dose can be assumed to be approximately proportional to the area of overlap of the films over the breast. In calculating the dose per examination it was assumed here that there was 100% overlap between films and this will have somewhat overestimated the true additional dose. Table 6 shows that some extra films are indeed taken where the larger format film is not available. For OB views about 7% of women required more than one film if the larger format film was not available. Even where the larger format film was available about 1.1% of women still required more than one film per oblique view. Fewer extra films were taken for the CC views, with only about 1.8% of women having extra films if the larger format film was not available. The effect of these differences on doses for examinations was compared in Table 7. It was calculated that there was about a 16% dose saving for single view screening if the larger format film was available. This dose saving fell to 9% for two view screening. Although the potential dose saving achievable by ensuring that all centres had large format film available would not be great for the screened population as a whole, it would be substantial for those women who currently need extra films. Women who had large compressed breast thicknesses were more likely to require extra films. Thus the dose reduction achieved by having the large format available is of most benefit to those women receiving the highest doses. Recent new equipment purchases for the NHSBSP have included funding for equipment to have facilities for both standard and large sizes of cassettes. Funding and radiation dose are not the only issues affecting the use of larger format cassettes. There may be image quality issues (e.g. relating to positioning, and effectiveness of compression) and practical ones (e.g. space for additional film handling equipment, film viewing arrangements). Nonetheless, more widespread use of large cassettes can be expected in future in the NHSBSP.
Standard breast dose
The MGD to the standard breast may be calculated relatively easily by combining the mAs for correct exposure of a 40 mm thick block of PMMA with values of tube output and beam quality [2]. The advantage of this approach is that the effect of equipment factors can be assessed, while eliminating patient variables. The exposure may be made using standard exposure factors or those selected automatically by the X-ray set. The standard breast doses shown here are for a tube voltage of 28 kV and a Mo/Mo target/filter combination. The limitation of the standard breast method is that it does not indicate the actual doses to individual women. The doses calculated for real breasts are for the beam qualities used clinically which can result in higher or lower doses than expected from the standard breast dose. In general one would expect the standard breast dose to be correlated with average breast doses for any mammography system. In Figure 7 the average MGD for 50–60 mm thick breasts and the MGD to standard breasts were linearly correlated with a coefficient of 0.54. Figure 7 also showed that the dose for an OB film on an average sized breast (i.e. 55 mm thick) was about 42% higher than the dose to the standard breast on the same system. This difference is largely explained by the fact that the standard breast is only 45 mm thick. For a few systems a very different relationship between the standard breast dose and the average breast dose was found. This is partly explained by the different beam qualities used. A further consideration is the inhomogeneous nature of breast tissue as compared with PMMA, and what effect that has on the operation of AEC systems.
In a recent review of the performance of mammographic equipment in the NHSBSP, the MGD for the standard breast was found to have gradually risen from 1.28 mGy in 1991 to 1.45 mGy in 2001 where a tube voltage of 28 kV and a Mo/Mo target/filter combination was used [17]. The average MGD to the standard breast was 1.55 mGy in 2001 where the clinically selected factors were used. These data for doses to the standard breast would lead one to expect a similar rise in the dose measurements for women attending screening. In fact this paper shows that such a trend has been counteracted by the changes in beam quality selection (e.g. increased use of rhodium filters). Such an apparent contradiction is one reason why a revision of how the MGD is estimated with phantoms is being considered in the UK.
Although the standard breast model has worked well for many years, some problems have become apparent recently and are illustrated by the data presented here. The first issue is that the standard breast model with a 45 mm thickness and 50% glandularity in the central region is not representative of the breasts of women attending for screening. In screening in the UK the average compressed breast has a thickness of approximately 50–60 mm and a glandularity of about 30% [5, 18]. This means that average doses are rather higher than the MGD to the standard breast for a given system. Another issue is the change in the beam quality selection by modern X-ray sets. A few years ago almost all mammography in the NHSBSP was conducted using a tube voltage of 28 kV and a Mo/Mo target/filter combination. Nowadays the beam quality is usually selected automatically depending on breast thickness and composition, and the spectrum used may have significantly higher or lower energies. As a result the use of a phantom comprising 40 mm PMMA results in an atypical spectrum being used to calculate the dose. To fully understand the doses delivered one needs to conduct dose measurements using either exposure data for real breast examinations or simulations with a range of thicknesses of PMMA. These procedures are time consuming and a more appropriate simple single measurement would be useful.
A procedure for measuring dose in mammography with a new standard breast model is being drafted [19]. It is proposed to use published data for the equivalence between PMMA and typical compressed breast to establish a new standard breast simulated with 45 mm PMMA [5]. The entrance air kerma for 45 mm PMMA is equivalent to that for a 53 mm thick breast with a glandularity of 29% in the central region. (Note that this composition was found to be approximately typical for breasts of this compressed thickness for women in the age range 50–64 years). This will bring the UK into line with existing European dose and quality control protocols, which have a standard PMMA phantom with a 45 mm thickness [20, 21]. It is planned to have a new dose recommendation whereby the MGD to the new standard breast model should not exceed 2.5 mGy when exposed using the beam quality selected clinically. In practice this will usually mean as selected automatically by the AEC with the phantom in place. This change in the standard breast model and the applicable dose limit is intended to make the standard dose measurement more relevant to the clinical situation, but have a neutral effect on the proportion of systems meeting the required standard.
Diagnostic reference level
Nine systems (3.5%) exceeded the diagnostic reference level of 3.5 mGy for an OB film of a 55 mm thick breast. Most of these X-ray sets were Lorad models using the old AEC software. The reasons for the high doses were a combination of a high mean glandular dose to the standard breast and a beam quality selection strategy that increased doses for this type of breast. The national DRL appears to be set at the correct level with only 3.5% failing in the NHSBSP where doses are relatively well controlled. It would be undesirable to set a level whereby a substantial proportion of systems failed, leading to dose reductions which might compromise image quality. The national DRL should be kept under review and reduced when this is possible without significantly impacting on image quality. Some screening centres may find that they can set a local DRL that is lower than the national DRL.
Film–screen speed
A major determinant of average breast doses was the type of film–screen combination used. All the film manufacturers offer a choice of screen speeds. Relatively few screening centres choose to use the faster screen options (Agfa Detail R and Detail S; Agfa-Gevaert, Mortsel, Belgium, Fuji UM medium; Fuji Photo Film, Bedford, UK; and Kodak Min-R2190; Eastman Kodak, Rochester, NY) because of the slight loss in image quality involved. The fastest film–screen combination was the Agfa HDR film used with an Agfa Detail R screen, which resulted in an average MGD to the standard breast of 0.96 mGy. The slowest film–screen combination was the Kodak Min-R2000 film used with a Kodak Min-R2000 screen, which resulted in an average MGD to the standard breast of 1.59 mGy. Figure 8 showed that multiplying the standard breast dose by the ratio of the average dose for OB films to the average dose to the standard breast (i.e. 1.57) was a very good predictor for the average doses for oblique films for a given film–screen combination. This was possible partly because film–screen speed is such a major determinant of dose. It must be remembered that the choice of film–screen combination affects not only the average doses but also the image quality.
Model of X-ray set
Another determinant of average breast doses is the model of X-ray set used. The average doses for different models of X-ray set varied from 1.54 mGy to 3.05 mGy as shown in Table 9. It was particularly noticeable that the Lorad Mark IV X-ray sets gave the highest doses. This was partly due to the design of the Bucky systems and partly due to the programming of the AEC beam quality selection software. The use of the new AEC software appears to be reducing doses by about 16%. Even with the new software, doses remain higher than for other models. The design factors that cause these differences in dose may or may not also affect image quality, depending on the underlying reasons for the differences.
High dose sub-group
Although the average dose per OB film is 2.2 mGy in the NHSBSP there are subgroups of women for whom the doses are higher. One identifiable sub-group of women who receive larger doses than average comprises those women with relatively thick breasts on compression. It was shown in Table 10 that the small sub-group of women with compressed breasts greater than 90 mm thick had doses for OB films of about 2.7 times the average for a given mammography system operated using manual beam quality selection. Another factor that affects the doses to women is the equipment used. This can be appreciated by examining the doses to the standard breast, which ranged from 0.76 mGy to 2.29 mGy. Although the quality assurance guidelines for the NHSBSP [3] require that the MGD to the standard breast should not exceed 2 mGy nine systems were slightly above this limit. For the mammography system with the highest standard breast dose of 2.29 mGy, breast doses of about 1.6 times the average can be expected. It is assumed here that doses to breasts of all sizes will be greater by approximately this amount. (One can see that this is likely to be true since a major cause of this variation is the variation in the speed of the film–screen system, which can be assumed to be approximately the same for all sizes of breast). Thus using this system one can expect that a few women with large breasts (>90 mm thick) would receive doses of about 10.7 mGy per film (i.e. 2.7 x 1.6 x 2.47 mGy) if beam quality were selected manually. In fact only 11 out of 32 250 oblique films (i.e. 0.03%) had doses in excess of 10.7 mGy. 99.97% of oblique films had doses of less than 10.7 mGy and this dose per OB film can be regarded as an upper limit to what can normally be expected for a screening mammogram in the NHSBSP. The doses for whole examinations were higher, particularly for 2V examinations. Using the same arguments as above and the data from Table 10 one can estimate the maximum dose for a CC film normally expected. For OB films, doses for the manual mode were used in this calculation as these resulted in higher doses. However there is a lack of data in Table 10 for compressed breasts over 90 mm thick in the CC view, and automatic mode is therefore used to estimate maximum doses. In this case the highest expected dose for a CC film is 5.8 mGy (i.e. 1.84 x 1.6 x 1.96 mGy). Thus the maximum dose that may be normally expected for a 2V examination is 16.5 mGy if there is one film per view. In practice only 15 women (0.16%) out of 9562 who had 2V examinations received more than 16.5 mGy even when all extra films were included. Thus one may conclude that a very small proportion of women will receive about 3.8 times (i.e. 16.5÷4.32) the average dose in a 2V screening programme, and that this identifiable sub-group should be considered in any risk benefit analysis. A few women will receive doses higher than this, but this is very rare and impossible to predict in advance. Note that some of the highest doses per examination are overestimates because multiple films per view have merely been added together – rather than by taking the amount of overlap into account.
|
Conclusions |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
|
Acknowledgments |
|---|
Received for publication April 19, 2004. Revision received September 6, 2004. Accepted for publication October 12, 2004.
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionsReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Berrington de Gonzalez, C. D. Berg, K. Visvanathan, and M. Robson Estimated Risk of Radiation-Induced Breast Cancer From Mammographic Screening for Young BRCA Mutation Carriers J Natl Cancer Inst, February 4, 2009; 101(3): 205 - 209. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. C. Oeffinger, J. S. Ford, C. S. Moskowitz, L. R. Diller, M. M. Hudson, J. F. Chou, S. M. Smith, A. C. Mertens, T. O. Henderson, D. L. Friedman, et al. Breast Cancer Surveillance Practices Among Women Previously Treated With Chest Radiation for a Childhood Cancer JAMA, January 28, 2009; 301(4): 404 - 414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Taylor and R. E. Taylor Surveillance for Breast Cancer After Childhood Cancer JAMA, January 28, 2009; 301(4): 435 - 436. [Full Text] [PDF]
|
|
|
|
 |
|
 F. A. Mettler Jr, W. Huda, T. T. Yoshizumi, and M. Mahesh Effective Doses in Radiology and Diagnostic Nuclear Medicine: A Catalog Radiology, July 1, 2008; 248(1): 254 - 263. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M ROBINSON and C J KOTRE Trends in compressed breast thickness and radiation dose in breast screening mammography Br. J. Radiol., March 1, 2008; 81(963): 214 - 218. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T Svahn, B Hemdal, M Ruschin, D P Chakraborty, I Andersson, A Tingberg, and S Mattsson Dose reduction and its influence on diagnostic accuracy and radiation risk in digital mammography: an observer performance study using an anthropomorphic breast phantom Br. J. Radiol., July 1, 2007; 80(955): 557 - 562. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K Faulkner, S McCormack, and K Bennison A retrospective analysis of digital stereotaxis in breast screening Br. J. Radiol., July 1, 2007; 80(955): 563 - 568. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J Law, K Faulkner, and K C Young Risk factors for induction of breast cancer by X-rays and their implications for breast screening Br. J. Radiol., April 1, 2007; 80(952): 261 - 266. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-H. Ng, N. Jamal, and L. DeWerd Global quality control perspective for the physical and technical aspects of screen-film mammography--image quality and radiation dose Radiat Prot Dosimetry, December 1, 2006; 121(4): 445 - 451. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K C Young, J M Oduko, H Bosmans, K Nijs, and L Martinez Optimal beam quality selection in digital mammography Br. J. Radiol., December 1, 2006; 79(948): 981 - 990. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C J Whitaker, C M Kelly, K Faulkner, and E C Stamp Influence of menopausal status and use of hormone replacement therapy on radiation dose from mammography in routine breast screening. Br. J. Radiol., July 1, 2006; 79(943): 597 - 602. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K C YOUNG Recent developments in digital mammography Imaging, June 1, 2006; 18(2): 68 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J Law and K Faulkner Radiation benefit and risk at the assessment stage of the UK Breast Screening Programme. Br. J. Radiol., June 1, 2006; 79(942): 479 - 482. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
BJR Review of the Year - 2005. Br. J. Radiol., March 1, 2006; 79(939): 183 - 187. [Full Text] [PDF]
|
|
|
|
|
|
J Law Breast dose from magnification films in mammography Br. J. Radiol., September 1, 2005; 78(933): 816 - 820. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
