British Journal of Radiology (2006) 79, 597-602
© 2006 British Institute of Radiology
doi: 10.1259/bjr/68819456
Influence of menopausal status and use of hormone replacement therapy on radiation dose from mammography in routine breast screening
C J Whitaker, BSc, PhD
C M Kelly, BA
K Faulkner, FIPEM, FInstP, FSRP
and
E C Stamp, MMathStat
Quality Assurance Reference Centre, 9 Kingfisher Way, Silverlink Business Park, Wallsend NE28 9ND, UK
Correspondence: Dr Keith Faulkner
 |
Abstract
|
|---|
Menopausal status and hormone replacement therapy (HRT) cause alterations in breast structure which can affect mammographic image quality. Here we present the results of a study to discover the effect of menopausal status and HRT use on breast dose. Women attending routine screening completed questionnaires which included questions regarding menopausal status and HRT use. Details of the radiographic technique factors were recorded, from which the mean glandular dose (MGD) per film for each woman was calculated. MGD values were analysed with regard to the woman's menopausal status and HRT use. The data from 516 women were analysed. Among the women who had never used HRT, women who had not undergone the menopause had a mean MGD of 2.94 mGy per film, whereas post-menopausal women had a lower mean MGD of 2.52 mGy per film: a difference which was found to be highly significant (p = 0.0045). Post-menopausal women who had never used HRT and those who had previously used HRT, but had ceased using it, had identical mean MGDs (2.54 mGy per film), whereas current HRT users had a significantly greater mean MGD (2.89 mGy per film, p = 0.003). Women currently using HRT receive a statistically significantly larger radiation dose from routine breast screening than other women. However, this effect is small and only occurs during the period of HRT use. Women who have ceased using HRT show no difference in MGD compared with women who have never taken HRT.
|
Introduction
|
|---|
In recent years, there has been increasing interest in the effects of hormone replacement therapy (HRT) on women's health [1–6]. HRT use has been increasing rapidly since its widespread introduction in the early 1970s [7, 8], with an estimated 50% of women currently aged 50 years and over having taken HRT at some time [9]. Two of the largest studies encompassing research into the effects of HRT on health are the Million Women Study (MWS) (recruitment: 1996–2001) in the UK [4, 9] and the Women's Health Initiative (WHI) (recruitment 1993–1998) in the USA [3, 5]. One of the main differences between these studies is that the WHI actively enrolled women into a clinical trial of HRT, whereas the MWS simply recorded the woman's HRT use, as prescribed by her GP.
The MWS was designed to investigate relationships between the patterns of use of HRT and certain medical conditions, including breast cancer, in women attending routine screening as part of the NHS Breast Screening Programme (NHSBSP) [4, 10]. Since the MWS began in 1996, it has recorded the personal details of over one million women [9], including specific details about previous medical conditions, family history of cancer, menopausal status and HRT use, by use of a self-administered questionnaire.
Although data analysis for the MWS is still ongoing, the study has already described some of the effects of HRT on breast screening and breast cancer [9, 11, 12]. As part of the MWS, women were tracked through the NHSBSP to determine rates of breast cancer, which were then analysed with regard to their original questionnaire details. It was found that current users of HRT were more likely to develop breast cancer than women who had never taken HRT and women who were past users of HRT [9]. Many other studies have also found a direct link between HRT use and increased incidence of breast cancer [1, 5, 13–15]. It was this HRT-induced increase in breast cancer which prompted the premature end to the WHI trial of combined oestrogen and progesterone HRT in 2002 [5].
With specific regard to the population studied in the MWS, i.e. women attending for mammograms as part of the NHSBSP, the direct effects of HRT on mammography need to be considered. This is because the clarity of a mammogram is affected by the tissue composition of the breast [16]. Fatty tissue has low absorption of X-rays, whereas ductal and glandular tissues absorb radiation and are, therefore, radiologically dense [17]. It is this ratio between fatty and dense breast tissue which determines overall breast density and which can be strongly influenced by both menopausal status and HRT use [16–19]. Despite this, the impact of menopausal status and HRT use on the radiation dose received by women during routine breast screening has not been established.
The objectives of this study were to:
- Determine the radiation dose received by 500 women during routine breast screening across five Breast Screening Units;
- Investigate the effect that a woman's menopausal status has on MGD from mammography;
- Establish whether taking HRT affects radiation dose and whether this effect continues after the cessation of HRT use;
- Compare the main types of HRT preparation to see what influence, if any, current HRT type has on mammographic radiation dose.
|
Methods and materials
|
|---|
Structure of the study
Women were recruited from five Breast Screening Units across the North East, Yorkshire and the Humber Government Office Regions of the UK between November 2002 and September 2003, with Units contributing between 13.2% and 31.6% of the women in the study (Table 1
). Ethical approval was obtained from the Northern and Yorkshire Multi-Centre Research Ethics Committee, as well as from each specific local Research Ethics Committee for each Unit.
Every woman who attended a particular screening session was given the opportunity to take part in the study. If the woman agreed, she was given an information sheet explaining the study and asked to complete a consent form and a modified version of the MWS questionnaire [20] whilst at the screening centre. In total, 520 women across the region gave their consent to the study and completed the questionnaire. The questionnaire was based on the MWS questionnaire [20] and included detailed questions on the woman's menopausal status and HRT use. However, in order that additional data on the woman could be obtained, the woman's individual NHS Screening Number (Sx number) was also recorded on each questionnaire.
Following the woman's routine mammogram, details of the radiographic technique used for each film (such as projection, tube potential, tube current exposure time product and breast thickness) were recorded. The radiographic data, along with the X-ray tube output measurements from the mammography machine, were inputted into the NHSBSP Breast Dose Calculator program [21], to estimate the mean glandular dose (MGD) for each film.
However, since the women had varying numbers of films taken each due to various factors including the introduction of two-view screening at three out of five of the Units in this study, the average MGD for each woman was calculated on a "per film" basis, giving each woman a mean MGD per film (mean MGD). There were only four women whose radiographic data was not sufficiently complete to calculate their MGD, leaving 516 women for inclusion in the data analysis.
Definition of groups for data analysis
The mean MGD value for each woman was linked to her questionnaire responses via her individual Sx number. This then allowed mean MGD values to be analysed with regard to the woman's menopausal status and details of her HRT use.
Initially, women were divided into three different menopausal groups, pre-, peri- and post-menopausal, based mainly on whether or not their periods had stopped at the time of screening. In addition, women aged 53 years and over who had had either a hysterectomy without oophorectomy (removal of both ovaries) or who had started taking HRT before their natural menopause were defined as post-menopausal, along with women of any age who had had a bilateral oophorectomy. This categorisation was in line with guidance from the Million Women Study [12]. Subsequently, the pre- and peri-menopausal women were combined for comparison against the mean MGD of post-menopausal women. All data were analysed using a one-way ANOVA to assess variation and to establish statistical significance.
As part of the questionnaire, women were asked whether they had ever taken HRT and, if so, whether they were currently taking HRT and which particular type. Their answers led to the post-menopausal women being categorised as never having taken HRT (never), previously having taken HRT (previous) or currently taking HRT (current). Mean MGD values were compared between the three groups and also for a combination of never and previous users, against current HRT users. Finally, current HRT users were divided according to the type of HRT they were taking; combined oestrogen and progesterone, oestrogen-only or another type of preparation. Again, the mean MGDs for the women in each group were compared using a one-way ANOVA.
|
Results
|
|---|
The characteristics of the women who took part in the study are outlined in Table 1
. The women ranged in age from 42 years to 82 years, with a mean age of 57.0 years, the majority being aged less than 60 years (72.7%). The vast majority of the women's questionnaires indicated that they were post-menopausal (88.0%), with the remainder being equally spilt between pre- and peri-menopausal (5.8% each). Only two women (0.4%) who completed the questionnaires failed to give details of their menopausal status.
Just under half of the women had taken HRT (44.9%) at some time, with 20.5% indicating that they were current users. As with menopausal status, only a small proportion of women (2.3%) did not specify whether they had ever taken HRT. For current HRT users, combined oestrogen and progesterone was taken by 42.5%, with oestrogen-only HRT comprising 29.2%. However, there was a large percentage of women (28.3%) who either used other forms of HRT (such as progesterone-only) or did not state the type of HRT on the questionnaire. This lack of knowledge about the type of HRT used may have led to difficulties in ascertaining any specific influence of the type of HRT on radiation dose (see Discussion).
The mean MGD data from 516 women were analysed using a one-way ANOVA. The mean MGD per film for each woman was analysed by assigning the women to a group depending on their menopausal status and details of their HRT use, as obtained from their questionnaires and defined in the Methods and Materials section.
Influence of menopausal status
Pre- and peri-menopausal women had similar mean MGD values of 2.84 mGy and 2.91 mGy, respectively (Figure 1). Post-menopausal women had a lower mean MGD of 2.62 mGy, although this difference was not statistically significant (p = 0.08). If, however, post-menopausal women who have never used HRT are compared directly with pre- and peri-menopausal women who have also never used HRT (Figure 2), then the difference in MGDs becomes statistically significant (p = 0.0045). These non-HRT post-menopausal women have a mean MGD of 2.52 mGy, compared with a mean MGD of 2.94 mGy for non-HRT pre- and peri-menopausal women.
View larger version (12K):
[in this window]
[in a new window]
|
Figure 1. Effect of menopausal status on mean glandular dose(MGD) per film. Error bars indicate 95% confidence levels from one-way ANOVA analysis.
|
|
View larger version (10K):
[in this window]
[in a new window]
|
Figure 2. Effect of hormone replacement therapy(HRT) use on mean glandular dose (MGD) per film of post-menopausal women (grey bars) and pre- and peri-menopausal women combined (dark grey bar). "Never" indicates women who have never used HRT, "Previous" indicates women who have taken HRT in the past, but have now ceased HRT use, and "Current" indicates women currently taking HRT. Error bars indicate 95% confidence levels from one-way ANOVA analysis.
|
|
Influence of HRT use
In post-menopausal women, women who had never used HRT and those women who had previously used HRT, but now did not, had practically identical mean MGDs of 2.54 mGy (Figure 2). However, current users of HRT had a greater mean MGD of 2.89 mGy, which was similar to the MGD for pre- and peri-menopausal women who had never taken HRT (2.91 mGy). This difference in MGD between current users of HRT and women not currently taking HRT was found to be highly statistically significantly (p = 0.003; Figure 3).
View larger version (9K):
[in this window]
[in a new window]
|
Figure 3. Effect of current hormone replacement therapy(HRT) use on the mean glandular dose (MGD) per film of post-menopausal women. Error bars indicate 95% confidence levels and *** indicates highly significant difference (p<0.005) from one-way ANOVA analysis.
|
|
Influence of HRT type
As described previously, the large proportion of women who did not specifically identify their current type of HRT in the questionnaire may have adversely affected the investigation as to whether the type of HRT influences MGD. For current HRT users who specified HRT type, there was no statistically significant difference in mean MGDs between combined (2.73 mGy) and oestrogen-only (2.88 mGy) HRT (Figure 4).
View larger version (11K):
[in this window]
[in a new window]
|
Figure 4. Effect of type of hormone replacement therapy(HRT) on the mean glandular dose (MGD) per film of current HRT users. "Combined" indicates combined oestrogen and progesterone users. Error bars indicate 95% confidence levels from one-way ANOVA analysis.
|
|
|
Discussion
|
|---|
The data obtained from the women's questionnaires was generally complete, with only 14 women (2.7%) not providing enough details to classify either their menopausal status or history of HRT use (Table 1). Overall, only four women (0.8%) were excluded from the data analysis due to lack of complete radiographic data.
The classification of women into three menopausal groups was carried out following the criteria used by the MWS [12]. The percentages of women in each category, pre-, peri- and post-menopausal, were 5.8%, 5.8% and 88% in the study reported here (Table 1) and 6.3%, 6.5% and 87% in the MWS [12], respectively. Classification of women based on HRT use was more straightforward than menopausal status and led to 45% of women being classified as a previous or current user of HRT. This figure is similar to the MWS where questionnaires indicated that almost half of women (47%) had used HRT at some time [4]. There is obviously close agreement between our study and the MWS with regard to the woman's menopausal details and HRT use.
In our study, there was a large proportion of current HRT users who were unable to accurately recall the name or type of HRT preparation they were currently taking. Combined oestrogen and progesterone accounted for 43% of current users, with oestrogen-only preparations accounting for 29% (Table 1). However, there was a large number of current HRT users classified as using either another or unknown type of HRT (28%). These figures are comparable with those obtained by the MWS. In the MWS, the majority of current HRT users took combined oestrogen and progesterone HRT (60%), with around a third taking oestrogen-only HRT (31%) and only 8.7% of women classed as other or unknown HRT type [9]. This large difference in the proportion of current HRT users classified as "unknown HRT type" between our study and the MWS is likely to be due to the difference in questionnaire protocol.
In the MWS, the women completed their questionnaire at home, where they were likely to have their current HRT to hand and so were able to identify the type accurately. However, since our questionnaire was completed at the screening unit, the women were less likely to be able to remember the exact name or type of preparation they were using. This led to difficulty in interpreting the radiation dose results with regard to HRT type for current HRT users. Although our results indicated a small (5%) difference in dose received between those women taking combined oestrogen and progesterone and those taking oestrogen-only HRT preparations (Figure 4), the lack of specific details about HRT type resulted in too few women to analyse the data meaningfully.
Radiation dose
The mean radiation dose received by the 516 women in this study was 2.66 mGy. This compares with an average MGD of 2.23 mGy determined by Young et al [22] for women attending routine breast screening across the UK during 2001 and 2002. The small difference between the two mean MGDs is likely to be due to the use of Lorad systems (Lorad M-IV) in four out of five of the Units which took part in our study. Lorad systems are known to give a higher MGD than other systems and, according to the Young et al study, account for only 15% of the systems used nationally [22].
Ideally, the best approach to estimating the glandular dose is to apply an accurate glandularity factor, specific to each woman. In the absence of detailed knowledge of breast glandularity for each woman, an average glandularity factor has been applied. As the glandularity between groups analysed here may vary, part of the observed differences between groups will be due to this effect. It is anticipated that the actual difference between groups will be smaller.
Influence of menopausal status on radiation dose
Although the mean MGD for post-menopausal women was around 9% lower than those for pre- and peri-menopausal women together, this difference was not statistically significant (p = 0.08) when HRT status was not taken into account, perhaps due to the HRT masking any real effect the menopause has on MGD. However, when women who had never taken HRT were analysed with regard to their menopausal status, the difference between this group of women and the post-menopausal women did reach statistical significance, with post-menopausal women being found to have received a statistically significantly smaller radiation dose (2.52 mGy) than women who had not yet undergone the menopause (2.94 mGy: p = 0.0045). The difference in radiation dose between these groups may reflect the changes which occur to the structure of a woman's breasts after the menopause [18].
Influence of HRT use on radiation dose
It was decided that only post-menopausal women would be analysed with regard to HRT use, since the numbers of pre- and peri-menopausal women were so small (60 in total) that it would severely limit any statistical analysis. It was also likely that the vast majority of non-menopausal women would never have taken HRT.
In post-menopausal women, current users of HRT received an approximately 12% larger radiation dose than women who were not taking HRT at present (Figure 2). This difference was highly statistically significant (p = 0.003). There was no statistical difference in radiation dose received between women who had stopped taking HRT and those who had never taken it (Figure 2). This temporary nature of the effect of HRT on radiation dose is supported by research which has shown that the effects of HRT use, both on the incidence of breast cancer [9, 14] and on the specificity of mammography [11], decline rapidly after initial cessation and return to the levels of women who have never used HRT. These results indicate that the influence of HRT on radiation dose, breast cancer and mammograms only lasts for the length of time that the woman is taking HRT, with little long term effect.
However, for current HRT users, the increased radiation dose received may have important implications. The difference between radiation dose received by a current user of HRT, versus a non-user is approximately 0.35 mGy per film, which equates to 0.7 mGy per routine two-view screen (four films). Although this additional amount of radiation received is relatively small, there may still be an impact when it comes to radiation-induced breast cancer. Law and Faulkner calculated the effect of specific radiation doses on the rates of induction of breast cancer [23]. For women aged 50–64 years, this equates to approximately 11.4 breast cancers per million women for each mGy of radiation.
The small amount of additional radiation incurred as a result of taking HRT is outweighed by the substantial benefit of screening women on HRT, since this group of women is known to have greater background incidence of breast cancer than women not taking HRT [1, 5, 9, 11–15]. Other factors, such as lack of specificity and selectivity of mammograms with HRT use [24, 25], are more likely to affect the benefit/risk ratio.
|
Conclusions
|
|---|
Women currently using HRT receive a statistically significantly larger radiation dose from routine breast screening than other women. However, this effect is small and only occurs during the period of HRT use. The increase in mammography radiation dose for women taking HRT may result in a small increase in the number of radiation-induced cancers, but this is probably justified due to the higher incidence of breast cancer in these women. Women who have ceased using HRT show no difference in radiation dose received compared with women who have never taken HRT.
This study was partially supported by the European Commission's Radiation Protection Research Programme, project DIMOND III (Measures for Optimising Radiological Information and Dose in Digital Imaging and Interventional Radiology), contract number FIGM-CT-2000-00061.
|
Acknowledgments
|
|---|
We wish to thank all the women who agreed to take part in the study and completed questionnaires, along with the staff at the five Breast Screening Programmes in the North East, Yorkshire and The Humber Government Office Region who were involved in this study. Thanks to Prof. V Beral, Ms B Crossley and Ms A Brown from the Cancer Research UK Epidemiology Unit, Oxford, for their initial advice about the questionnaire and subsequent computerized data extraction from the questionnaires. The authors are also grateful to Mrs C Clayton, Mr G McGill, Mr C Moore and Dr K J Robson for their help with the Medical Physics data.
Received for publication September 1, 2005.
Revision received October 6, 2005.
Accepted for publication October 25, 2005.
|
References
|
|---|
- Goddard MK. Hormone replacement therapy and breast cancer; endometrial cancer and cardiovascular disease: risks and benefits. Br J Gen Prac 1992;42:120–5.[Medline]
- Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA 1998;280:605–13.[Abstract/Free Full Text]
- Women's Health Initiative Study Group. Design of the Women's Health Initiative Clinical Trial and Observational Study. Control Clin Trials 1998;19:61–109.[CrossRef][Medline]
- The Million Women Study Collaborative Group: The Million Women Study: design and characteristics of the study population [peer-reviewed research]
- Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women. Principal results from the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321–33.[Abstract/Free Full Text]
- Li C, Wilawan K, Samsioe G, Lidfeldt J, Agardh C-D, Nerbrand C. Health profile of middle-aged women: the Women's Health in the Lund Area (WHILA) study. Human Repro 2002;17:1379–85.[Abstract/Free Full Text]
- Gambrell RD Jr. Studies of endometrial and breast disease with hormone replacement therapy. In: Studd JWW, Whitehead MI, editors. The menopause. Oxford, UK: Blackwell Scientific Publications, 1988:247–61
- Jolleys JV, Olesen F. A comparative study of prescribing of hormone replacement therapy in USA and Europe. Maturitas 1996;23:47–53.[CrossRef][Medline]
- Million Women Study Collaborators. Breast cancer and hormone-replacement therapy in the Million Women Study. Lancet 2003;362:419–27.[CrossRef][Medline]
- http://cancerscreening.org.uk/breastscreen/ [Accessed 2 February 2006]
- Banks E, Reeves G, Beral V, Bull D, Crossley B, Simmonds M, et al. Impact of use of hormone replacement therapy on false positive recall in the NHS breast screening programme: results from the Million Women Study. BMJ 2004;328:1291–2.[Free Full Text]
- Banks E, Reeves G, Beral V, Bull D, Crossley B, Simmonds M, et al. Influence of personal characteristics of individual women on sensitivity and specificity of mammography in the Million Women Study: cohort study. BMJ 2004;329:477[Abstract/Free Full Text]
- Marsden J. Hormone-replacement therapy and breast cancer. Lancet Oncol 2002;3:303–11.[CrossRef][Medline]
- Coombs NJ, Taylor R, Wilcken N, Boyages J. Hormone replacement therapy and breast cancer: estimate of risk. BMJ 2005;331:347–9.[Free Full Text]
- Ewertz M, Mellemkjaer L, Poulsen AH, Friis S, Sorensen HT, Pedersen L, et al. Hormone use for menopausal symptoms and risk of breast cancer. A Danish cohort study. Br J Cancer 2005;92:1293–7.[Medline]
- Warren R. Hormones and mammographic breast density. Maturitas 2004;49:67–78.[CrossRef][Medline]
- Sterns EE, Zee B. Mammographic density changes in perimenopausal and postmenopausal women: is effect of hormone replacement therapy predictable? Breast Cancer Res Treat 2000;59:125–32.[CrossRef][Medline]
- Miller WR, Anderson TJ. Oestrogens, progestogens and the breast. In: Studd JWW, Whitehead MI, editors. The menopause. Oxford, UK: Blackwell Scientific Publications, 1988:234–46
- Greendale GA, Reboussin BA, Sie A, Singh R, Olson LK, Gatewood O, et al. Effects of estrogen and estrogen-progestin on mammographic parenchymal density. Ann Intern Med 1999;130:262–9.[Abstract/Free Full Text]
- http://www.millionwomenstudy.org/[Accessed 2 February 2006]
- http://www.cancerscreening.nhs.uk/breastscreen/publications/mammography-equipment.html#me-iqad [Accessed 2 February 2006]
- Young KC, Burch A, Oduko JM. Radiation doses received in the UK Breast Screening Programme in 2001 and 2002. Br J Radiol 2005;78:207–18.[Abstract/Free Full Text]
- Law J, Faulkner K. Cancers detected and induced, and associated risk and benefit, in a breast screening programme. Br J Radiol 2001;74:1121–7.[Abstract/Free Full Text]
- Evans A. Hormone replacement therapy and mammographic screening. Clin Radiol 2002;57:563–4.[CrossRef][Medline]
- Carney PA, Miglioretti DL, Yankaskas BC, Lerlikowske K, Rosenberg R, Rutter CM, et al. Individual and combined affects of age, breast density, and hormone replacement therapy use on the accuracy of screening mammography. Ann Intern Med 2003;138:168–75.[Abstract/Free Full Text]
Top
Abstract
Introduction
