Atlas of Mammography
Anatomy of the Breast
The breast or mammary gland is a modified sweat gland that has the specific function of milk production. An understanding of the basic anatomy, physiology, and histology is important in the interpretation of mammography. With an understanding of the normal breast, one is better able to correlate radiologic-pathologic entities.
The development of the breast begins in the fifth-week embryo with the formation of the primitive milk streak from axilla to groin. The band develops into the mammary ridge in the thoracic area and regresses elsewhere.
If there is incomplete regression or dispersion of the milk streak, there is accessory mammary tissue present in the adult, which occurs in 2% to 6% of women (1). Accessory breast tissue, particularly in the axillary area, that is separate from the bulk of the parenchyma may be identified on mammography in these women (2) (Fig. 1.1). The orientation of the milk streak is slightly lateral to the nipple above the nipple line and medial to the nipple below the nipple line. Therefore, patients with accessory breasts, accessory parenchyma, or accessory nipples are found to have these in the axillary region or just medial to the nipple in the inferior aspect of the breast or upper abdominal wall (Figs. 1.2, 1.3). In women with accessory breast tissue, changes occur cyclically and with pregnancy and lactation, as they do within the breasts themselves. Therefore, these patients may note that areas of accessory breast tissue may enlarge with pregnancy and may produce milk when the patient is lactating if there is a duct orifice or nipple present.
At 7 to 8 weeks of embryologic development, there is an invagination into the mesenchyma of the chest wall. Mesenchymal cells differentiate into the smooth muscle of the nipple and areola (1,3). At 16 weeks, epithelial buds develop and branch. Between 20 and 32 weeks, placental sex hormones entering the fetal circulation induce canalization of the epithelial buds to form the mammary ducts. At 32 to 40 weeks, differentiation of the parenchyma occurs, with the formation of the lobules (3,4).
The mammary gland mass increases by fourfold, and the nipple-areolar complex develops (1). Developmental anomalies include polymastia (accessory breasts along the milk streak), polythelia (accessory nipples), hypoplasia of the breast (Fig. 1.4), amastia (absence of the breast), and amazia (absence of breast parenchyma) (Fig. 1.5) (1). Systemic or iatrogenic influences in childhood may be related to breast hypoplasia or amazia. Iatrogenic causes of amazia include excision of the breast bud during biopsy of the prepubertal breast and the use of radiation therapy to the chest wall during childhood (1) (Fig. 1.6).
During puberty in girls, the release of follicle-stimulating hormone and luteinizing hormone by the pituitary causes release of estrogens by the ovary. Hormonal stimulation induces growth and maturation of the breasts. In early adolescence, the estrogen synthesis by the ovary predominates over progesterone synthesis. The physiologic effect of estrogen on the developing breast is to stimulate longitudinal ductal growth and the formation of terminal ductule buds (1). Periductal connective tissue and fat deposition increase (1), accounting for increase in size and density of the breasts.
The adult breast is composed of three basic structures: the skin, the subcutaneous fat, and the breast tissue, which includes the parenchyma and the stroma. Beneath the breast is the pectoralis major muscle, which is also imaged during mammography. The breast parenchyma is enveloped by deep and superficial fascial layers; Cooper's ligaments, the fibrous strands that support the breasts, traverse the parenchyma and attach to the fascial layers. The parenchyma is divided into 15 to 20 segments, with each drained by a lactiferous duct (Fig. 1.7). The lactiferous ducts converge beneath the nipple, with about 5 to 10 major ducts draining into the nipple. Each duct drains a lobe composed of 20 to 40 lobules (Fig. 1.8).
Figure 1.1 HISTORY: A 30-year-old woman in the 32nd week of pregnancy, presenting with an enlarging axillary mass.
MAMMOGRAPHY: Right axillary (A) view shows a prominent ductal and glandular pattern in the area of the mass in the axilla. On ultrasound (B), dilated ducts are noted in the subcutaneous area. The findings are consistent with accessory breast tissue that is enlarging secondary to the pregnancy.
IMPRESSION: Accessory breast in the axilla, with changes related to pregnancy.
The microanatomy of the breast was described by Parks in 1959 (4). Each lobule is 1 to 2 mm in diameter and contains a complex system of tiny ducts (the ductules), which terminate in blind endings. The ductules can respond to hormonal stimulation of pregnancy by proliferation and formation of alveoli (3). Two types of stroma are present: the perilobular connective tissue, which contains collagen and fat, and the intralobular connective tissue, which does not contain fat (4).
Wellings et al. (5) further classified the microstructure of the normal breast into the terminal duct lobular unit (TDLU) (Fig 1.9). Small branches of the lactiferous ducts lead into terminal ducts that drain a single lobule. The terminal duct is composed of the extralobular segment and the intralobular segment. The lobule is composed of the intralobular terminal duct and the blindly ending ductules (5). The ductules are lined by a single layer of epithelial cells and a flattened peripheral layer of myoepithelial cells (5). A loose fibrous connective tissue stroma supports the ductules of the lobule.
The TDLU is a hormone-sensitive gland varying from 1 to 8 mm in diameter in the nonpregnant state and having the potential of milk production (6). The lobules normally regress at menopause, leaving blunt terminal ducts; however, in women older than 55 years with breast cancer, Jensen et al. (6) found the TDLUs remain well developed.
The work of Wellings et al. (5) has suggested that the TDLU is a basic histopathologic unit of breast from which many benign and malignant lesions arise. Fibroadenomas, sclerosing adenosis, apocrine cysts, lobular hyperplasia, and lobular carcinoma in situ are thought to develop in the lobule itself; ductal hyperplasia and ductal carcinoma in situ develop in the TDLU. Solitary intraductal papillomas, epithelial hyperplasia of the larger ducts, and duct ectasia occur in the main lactiferous ducts (4). Correlative studies between radiographic and histologic appearances of the breast parenchyma suggest that small nodular densities on mammography represent lesions of the terminal duct lobular units and that linear densities are due to periductal and perilobular fibrosis (7).
Blood Supply and Lymphatic Drainage
The primary arterial supply to the breast is from the perforating branches of the internal mammary and lateral
thoracic arteries. Minor contributions to the blood come from the branches of the thoracoacromial, subscapular, and thoracodorsal arteries (1). Venous drainage is primarily via branches of the internal mammary, intercostal, and axillary veins. If there is obstruction of the subclavian vein, collateral drainage produces dilated, tortuous vascular structures, easily visible on mammography (Fig. 1.10).
Figure 1.2 HISTORY: A 50-year-old woman with fullness in the right breast inferiorly.
MAMMOGRAPHY: Bilateral MLO (A) and CC (B) views show heterogeneously dense breasts. In the right breast at 6 o'clock is a prominent focal area of asymmetric breast tissue (arrow). Ultrasound showed no focal abnormality. No mass was palpable on clinical examination of this area.
IMPRESSION: Focal asymmetric breast tissue consistent with accessory breast.
NOTE: Accessory breast tissue is typically located laterally above the nipple line and medially below the nipple line.
Figure 1.3 HISTORY: A 55-year-old woman with slight fullness of the inferior aspect of the right breast.
MAMMOGRAPHY: Right CC (A) and MLO (B) views show a focal rounded asymmetry in the inferior aspect of the right breast, located slightly medially (arrow). On a ML (C) view of the lower aspect of the breast, the very inferior location of the asymmetry in the inframammary area is noted. On the cleavage view (D), the rounded aspect of the asymmetry is seen.
IMPRESSION: Accessory breast tissue.
NOTE: Accessory breast tissue below the nipple line is located slightly medially, because it develops from the primitive milk streak.
Figure 1.4 HISTORY: A 26-year-old gravida 0, para 0 woman with a history of ectodermal dysplasia. She had bilateral breast implants placed during adolescence because of the lack of breast development.
MAMMOGRAPHY: Bilateral MLO views. There are bilateral breast implants present. There is some dense glandular tissue in the subareolar area on the left side, but there are only rudimentary ducts on the right (arrow). The appearance on the right is similar to a normal male breast or a preadolescent female breast. In the condition of ectodermal dysplasia, there is a lack of normal development of epithelial structures such as nails, teeth, skin, hair, and sweat glands. Because the breast is a modified sweat gland and is derived from epithelium, the development of the breast can be impaired in this condition.
IMPRESSION: Maldevelopment of the breast secondary to ectodermal dysplasia.
Figure 1.5 HISTORY: Screening mammogram in a patient who is status post augmentation mammoplasty that was performed for marked asymmetry of breast size.
MAMMOGRAPHY: Bilateral MLO (A) and MLO implant displaced (B) views show that prepectoral saline implants are present. On the displaced views (B), the left pectoralis major muscle is present, but no similar structure is seen on the right. There is marked disparity of breast size, with the right breast being smaller and less glandular than the left, also confirmed on the CC implant displaced (C) views.
IMPRESSION: Mammary hypoplasia secondary to Poland's syndrome.
NOTE: Poland's syndrome is lack of development of the pectoralis major muscle.
Lymphatic drainage is via the superficial plexus to the deep plexus to the axillary and internal mammary lymph nodes. The low axillary nodes are often visible on mammography, as are small intramammary nodes. It is unusual to identify on mammography intramammary nodes in a location other than the superficial region of the middle- to upper-outer quadrant of the breast.
The breast lays over the musculature of the chest wall: the pectoralis major and minor muscles. The pectoralis major muscle has its origins at the anterior medial surface of the clavicle, the sternum, and the aponeurosis of the external oblique muscle and its insertion on the proximal humerus. The pectoralis major muscle, therefore, lies obliquely over the chest wall. This angle of obliquity varies with the body type of the individual. A parameter for a well-positioned mediolateral oblique (MLO) mammographic view is that the pectoralis major muscle is visible from the axilla down to the level of the nipple. Before positioning the patient for the MLO view, the
technologist must determine the angle of obliquity of the pectoralis major muscle. She then angles the mammographic receptor and compression device into this position. She is thereby able to compress the breast along the plane of the pectoralis muscle and to include more breast tissue. The pectoralis major muscle may be seen on the craniocaudal (CC) view in about one fourth to one third patients. Often, the muscle is seen as an area of density along the posterior aspect of the breast. Occasionally, the medial aspect of the muscle at the sternal border is prominent and can appear masslike on the CC view only.
Figure 1.6 HISTORY: A 49-year-old gravida 2, para 2 woman with a history of a plasma cell tumor of the left lung, treated with pneumonectomy and radiation therapy at age 4 years. The left breast has been significantly smaller than the right since development. The patient has no history of breast surgery.
MAMMOGRAPHY: Bilateral MLO (A) and CC (B) views. Marked asymmetry in the appearance of the breasts is seen. The left breast is significantly smaller, and there is a paucity of glandular tissue in comparison with the right. This striking lack of glandular development is presumably related to the lack of development of the breast bud, either from atrophy secondary to the radiation therapy or from surgery in the left midchest area, which may have involved incidental removal of part of the breast bud.
IMPRESSION: Hypoplasia of the left breast, presumably of iatrogenic origin.
Figure 1.7 Gross anatomy of the normal breast.
An inconstant muscle that may be present unilaterally or bilaterally is the sternalis muscle. This is a muscle band that runs vertically, parallel to the sternum. The sternalis muscle occurs in 3% to 5% of individuals and is more frequently observed in women. When present, the sternalis may be visible on mammography as a triangular or rounded density on the CC view only, and it is located at the medial, posterior edge of the breast (Figs. 1.11,1.12,1.13,1.14). It is not evident on the MLO or mediolateral (ML) views, and ultrasound is normal. If in doubt, a
computed tomography (CT) scan can be performed and will demonstrate the sternalis muscle clearly.
Figure 1.8 HISTORY: Patient presenting with a left serous nipple discharge.
GALACTOGRAM: Left CC (A) and ML (B) views show filling of the parenchymal system in the upper-inner quadrant via the cannulated duct. The normal ductal structures are seen, ramifying back into smaller ductal elements and eventually filling the rounded lobules.
IMPRESSION: Normal ductal anatomy.
Figure 1.9 Classification of the microstructure of the breast (from
Wellings SR, Jensen HM, Marcum RG. An atlas of subgross pathology of the human breast with special reference to precancerous lesions. J Natl Cancer Inst 1975;55:231–273.
At birth and in childhood, only rudimentary ducts are present (Fig. 1.15). At puberty, growth and elongation of ducts occur, and buds of the future lobules form at the end of the ducts (4,8). Periductal collagen is deposited, and mammographically the breast appears very dense and homogeneous. In the adult, in response to progesterone, the second stage of glandular development occurs, namely, the formation of the lobules (8) (Fig. 1.16).
With pregnancy, changes in the parenchyma occur to make milk secretion possible (8). There is marked increase in numbers of lobules and an increase in their size and complexity (4) (Fig. 1.17). In the second and third trimesters of pregnancy, the terminal ductules expand into the secreting alveoli (4). Prolactin, in the presence of insulin, growth hormone, and cortisol, changes the epithelial cells of the alveoli into a secretory state (1). With
the onset of lactation, the alveoli become maximally dilated, and milk production occurs. On mammography (Figs. 1.18 and 1.19), the lactating breast usually appears extremely dense, and dilated ducts may be seen. However, in a study of 18 women who were pregnant or lactating, Swinford et al. (9) found that these patients did not always have dense breasts. In this study, all seven lactating women had heterogeneously dense or extensively dense breasts, yet in 4 of 6 women, the density had not increased from the prepartum mammogram. After lactation ceases, the hypertrophied lobules shrink and may disappear. The breasts of parous women tend to appear more fatty and radiolucent than those of nulliparous women.
Figure 1.10 HISTORY: A 71-year-old woman with a history of diabetes, chronic renal failure, pulmonary embolism, and thrombophilia.
MAMMOGRAPHY: Bilateral CC (A) and MLO (B) views show circuitous vascular structures extending into the axillary regions. These are enlarged venous collaterals, likely secondary to the history of pulmonary embolism and a clot in the superior vena cava.
IMPRESSION: Enlarged venous collaterals.
Figure 1.11 HISTORY: A 51-year-old women for screening mammography.
MAMMOGRAPHY: Bilateral MLO (A) and CC (B) views show scattered fibroglandular densities. There are bilateral dense ovoid masslike densities located far medially at the chest wall (arrows). The obtuse angle at the chest wall is typical of a muscular structure.
IMPRESSION: Sternalis muscle.
Figure 1.12 HISTORY: Screening mammogram.
MAMMOGRAPHY: Bilateral CC views (A) show a focal masslike density in the far medial posterior aspect of the left breast (arrow). The normal pectoralis major muscle extends obliquely down the chest wall on the MLO view (B), however, the focal density is not seen. The masslike density is evident on the cleavage view (C). On the rolled CC lateral (D) and medial (E) views, the density changes shape and appears more obtuse at the chest wall.
IMPRESSION: Sternalis muscle.
With menopause, there is further involution of the parenchyma. The terminal lobules disappear, and the small ducts eventually atrophy. The main ducts are not greatly affected (4) (Fig. 1.20). The postmenopausal breast appears more radiolucent (10,11), and only minimal glandular elements are generally seen (Fig. 1.21). However, in some patients who have marked fibrocystic changes premenopausally or who are nulliparous, persistent dense parenchyma may be seen postmenopausally (Fig. 1.22).
Figure 1.13 HISTORY: Screening mammogram.
MAMMOGRAPHY: Right CC view (A) shows a focal density posteriorly in the medial aspect of the right breast (arrow) not seen on the MLO (B) view. The density is more triangular in appearance on the exaggerated CC medial (C) view; no abnormality was found on ultrasound (D). The location and appearance of this structure are characteristic of a normal variant, the sternalis muscle.
IMPRESSION: Sternalis muscle.
The effects of endogenous hormones related to the menstrual cycle have been observed on the histologic appearance of the normal breast (12). During the first half of the menstrual cycle, the effect of estrogen is to stimulate breast epithelial proliferation. In the second half of the cycle, after ovulation, progesterone causes ductal dilatation and differentiation of the ductular epithelial cells into secretory cells. In the 3 to 4 days before menses, edema and enhanced ductular acinar proliferation occur (1,12,13). Mammography at this time is more difficult
because the breasts are tender and may appear more dense. In a study of women aged 40 to 49 years, who were not on exogenous hormones, White et al. (14) found that breast parenchyma is less radiographically dense in the follicular rather than the luteal phase of the menstrual cycle. Postmenstrually, the edema is reduced, and secretory activity of the epithelium regresses (1,12).
Figure 1.14 HISTORY: A 46-year-old patient referred for needle localization of a right breast mass.
MAMMOGRAPHY: Right CC view (A) shows an oval masslike density located at the chest wall, far posteriorly (arrow). The lesion was not seen on the MLO view (B), however, it persisted on the spot-compression CC (C) view. On the rolled medial CC view (D), the area changes shape considerably, becoming more obtuse with the chest wall. Ultrasound was performed of the entire medial aspect of the breast and was negative. On CT (E), the asymmetry of the parasternal musculature is seen (arrow). There is an accessory muscle on the right consistent with sternalis muscle and corresponding to the mammographic finding.
IMPRESSION: Right sternalis muscle.
Figure 1.15 Changes that the normal breast undergoes during the life cycle.
Exogenous hormones may also have an effect on the mammographic appearance of the breast (Fig. 1.23). An increase in mammography density has been observed in 10% to 73% of women on combined therapy with estrogen and progesterone (15,16,17,18,19,20,21,22,23,24). Marugg et al. (19) found that 31% of patients treated with combination hormone replacement therapy (HRT) had an increase in fibroglandular tissue compared with 8.7% of women treated with estrogens alone, and this difference was statistically significant. Laya et al. (20) found that the increase in density was more pronounced in women with a lower baseline
parenchymal density. In the Women's Health Initiative randomized trial, McTiernan et al. (25) found that the mean mammographic percent density increased by 6.0% at year 1 in women placed on combined HRT versus decreasing by 0.9% in women not on hormones. Rutter et al. (26) found that the relative risk of an increase in breast density was 2.57 in women who initiated HRT; after discontinuation of HRT, the relative risk of a density decrease was 1.81. Therefore, the breast density changes associated with HRT are dynamic, increasing with initiation and decreasing with discontinuation.
Figure 1.16 HISTORY: A 32-year-old gravida 2, para 2 woman with a positive family history of breast cancer, for screening mammography.
MAMMOGRAPHY: Bilateral MLO (A) and CL (B) views show diffusely dense breast tissue with no focal abnormalities. This parenchymal pattern is often seen in young patients, and the density is related to parenchymal and stromal elements.
IMPRESSION: Normal breast, young patient.
Figure 1.17 HISTOPATHOLOGY: High power section of a lactating breast: the lobules are distended and filled with an eosinophilic material; secretory vacuoles are noted in the cells lining the glands.
Figure 1.18 HISTORY: A 32-year-old lactating woman, with a family history of premenopausal breast cancer in the grandmother, presents with a palpable right subareolar mass.
MAMMOGRAPHY: Bilateral CC (A) views show marked increase in density bilaterally with ductal dilatation. Comparison with a prior study (B) shows the parenchymal changes related to lactation. Ultrasound of the subareolar area (C) shows fluid-filled dilated ducts.
IMPRESSION: Lactational changes, dilated ducts.
Figure 1.19 HISTORY: Screening mammograms 1 year apart in a 37-year-old patient at high risk for breast cancer. At the time of the second study, the patient was lactating.
MAMMOGRAPHY: Initial left MLO (A) and CC (B) views show heterogeneously dense parenchyma. On the subsequent postpartum study(C, D), when the patient was lactating, there was marked overall increase in density and size of the breast. These changes were bilateral and are consistent with the normal lactating breast.
IMPRESSION: Normal changes in the lactating breast.
Stomper et al. (16) found changes related to hormonal influences in 24% of women placed on estrogen; these changes included diffuse increase in density (14%), multifocal areas of asymmetry (4%), and cyst formation (6%). Trapido et al. (27) found an increased risk of benign breast disease (both fibroadenomas and fibrocystic disease) in women on estrogen replacement therapy in comparison with a control group. The risk of benign breast disease was greater with increasing years of use of estrogen and was also higher in women with bilateral oophorectomy than in other postmenopausal women (27).
Another systemic effect on the breast, weight loss, may produce a striking change in the mammogram. The loss of body fat is accompanied by a loss of fat in the breasts, and the density of the parenchyma may appear much greater on mammography. Similarly, weight gain may cause the breasts to appear less radiographically dense because of increased fat deposition.
Danazol is sometimes used in the treatment of severe fibrocystic and cystic disease of the breast. The effect of danazol on the breast is to decrease pain and tenderness. The density of the breast also may decrease on the mammogram, allowing better visualization of the parenchyma (28). Tobiassen et al. (29) found that a significant decrease in the amount of glandular tissue on mammography was present in women treated with danazol for fibrocystic changes. The mammographic visualization of cysts
increased initially because of the marked regression in the obscuring glandular tissue, but thereafter, a decrease in the size and number of cysts occurred.
Figure 1.20 HISTORY: A 62-year-old woman for screening.
MAMMOGRAPHY: Bilateral MLO (A) and CC (B) views show scattered fibroglandular densities. There are prominent tubular densities in both subareolar areas, radiating back from the nipple.
IMPRESSION: Bilateral duct ectasia.
NOTE: Because the ducts are evident as discrete tubular structures, and because of their diameter, duct ectasia is present.
Figure 1.21 HISTORY: A 66-year-old para 0 woman for screening mammography.
MAMMOGRAPHY: Bilateral MLO (A) and CC (B) views show the breasts to be fatty replaced. These findings are the typical mammographic findings in a postmenopausal patient without a history of fibrocystic changes.
IMPRESSION: Normal postmenopausal breasts.
Figure 1.22 HISTORY: A 62-year-old gravida 2, para 2 patient for screening mammography. The patient was not taking hormone replacement therapy. She had a history of fibrocystic breasts.
MAMMOGRAPHY: Bilateral MLO (A) and CC (B) views show heterogeneously dense symmetrical-appearing breasts. Although the patient was not on hormones, the parenchyma is dense for the age of the patient. These findings often reflect diffuse fibrocystic changes.
IMPRESSION: Dense breasts in a postmenopausal patient.
Figure 1.23 HISTORY: Screening mammograms a year apart in a postmenopausal patient who was placed on hormone replacement therapy in the interval.
MAMMOGRAPHY: Initial bilateral MLO (A) views show scattered fibroglandular densities. One year later (B), there is marked overall increase in parenchymal density bilaterally, consistent with changes related to hormone replacement therapy.
IMPRESSION: Effect of hormone replacement therapy on mammography.
A decrease in breast density has been described in premenopausal women who take calcium and vitamin D supplements. Berube et al. (30) found that a total daily intake of 400 IU vitamin D and 1,000 mg calcium was associated with an 8.5% lower mean breast density in premenopausal women. A similar effect was not observed in postmenopausal women.
The effect of tamoxifen on glandular density has been studied in several small series. Konez et al. (31) found that the majority of patients on tamoxifen therapy had no change in parenchymal density on mammography. Tiersten et al. (32) found that breast density was inversely related to age and postmenopausal status, but that there was no correlation between breast density and tamoxifen use. However, Chow et al. (33), in a study of 28 high-risk women on tamoxifen, did find a decrease in mammographic density as quantitated on digitized images.
By keeping in mind the normal structure of the breast—both macroscopically and microscopically—and the effects on the breast of the hormonal changes during the life cycle, one can be better prepared to interpret the normal and the abnormal mammogram.
- Osborne M. Breast development and anatomy. In: Harris JR, et al. Breast Disease.Philadelphia: JB Lippincott, 1987;1–14.
- Adler DD, Rebner M, Pennes DR. Accessory breast tissue in the axilla: mammography appearance. Radiology1987;163: 709–711.
- Hughes ESR. The development of the mammary gland. Ann R Coll Surg Engl1950;6:99–105.
- Parks AG. The micro-anatomy of the breast. Ann R Coll Surg Engl1959;25:235–251.
- Wellings SR, Jensen HM, Marcum RG. An atlas of subgross pathology of the human breast with special reference to precancerous lesions. J Natl Cancer Inst1975;55:231–273.
- Jensen HM. On the origin and progression of human breast cancer. Am J Obstet Gynecol1986;154:1280–1284.
- Wellings SR, Wolfe JN. Correlation studies of the histological and radiographic appearance of the breast parenchyma. Radiology1978;129:299–306.
- Netter F. The reproductive system. In: Oppenheimer E, ed. The CIBA Collection of Medical Illustrations.Summit, NJ: CIBA, 1965:2.
- Swinford AE, Adler DD, Garver KA. Mammographic appearance of the breasts during pregnancy and lactation: false assumptions. Acad Radiol1998;5(7):567–572.
- Fewins HE, Whitehouse GH, Leinster SJ. Changes in breast parenchymal patterns with increasing age. Breast Dis1990;3: 145–151.
- Boyd N, Martin L, Stone J, et al. A longitudinal study of the effects of menopause on mammographic features. Cancer Epidemiol Biomarkers Prev2002;11(10 Pt 1):1048–1053.
- Vogel PM, Georgiade NG, Fetter BF, et al. The correlation of histologic changes in the human breast with the menstrual cycle. Am J Pathol1981;104:23–34.
- Fanger H, Ree HJ. Cyclic changes of human mammary gland epithelium in relation to the menstrual cycle: an ultrastructural study.Cancer1974;34:574–585.
- White E, Velentgas P, Mandelson MT, et al. Variation in mammographic breast density by time in menstrual cycle among women aged 40–49 years. J Natl Cancer Inst1998;90(12): 875–876.
- Pock DR, Lowman RM. Estrogen and the postmenopausal breast. JAMA1978;240:1733–1735.
- Stomper PC, Vanvoorhis BJ, Ravnikar VA, et al. Mammographic changes associated with postmenopausal hormone replacement therapy: a longitudinal study. Radiology1990; 174:487–490.
- Marchesoni D, Driul L, Ianni A, et al. Postmenopausal hormone therapy and mammographic breast density. Maturitas2006;53(1):59–64.
- Conner P, Svane G, Azavedo E, et al. Mammographic breast density, hormones, and growth factors during continuous combined hormone therapy. Fertil Steril2004;81(6):1617–1623.
- Marugg RC, van der Mooren MJ, Hendriks JH, et al. Mammographic changes in postmenopausal women on hormonal replacement therapy. Eur Radiol1997;7(5):749–755.
- Laya MB, Larson EB, Taplin SH, et al. Effect of estrogen replacement therapy on the specificity and sensitivity of screening mammography. J Natl Cancer Inst1996;88(10):627–628.
- Jackson VP, San Martin JA, Secrest RJ, et al. Comparison of the effect of raloxifene and continuous-combined hormone therapy on mammographic breast density and breast tenderness in postmenopausal women. Am J Obstet Gynecol2003;188(2): 389–394.
- Persson I, Thurfjell E, Holmberg L. Effect of estrogen and estrogen-progestin replacement regimens on mammographic breast parenchymal density. J Clin Oncol1997;15(1):3201–3207.
- Berkowitz JE, Gatewood OMB, Goldblum LE, et al. Hormonal replacement therapy: mammographic manifestations. Radiology1990;174:199–201.
- McNicholas MMJ, Heneghan JP, Milner MH, et al. Pain and increased mammographic density in women receiving hormone replacement therapy: a prospective study. AJR Am J Roentgenol1994;163:311–315.
- McTiernan A, Martin CF, Peck JD, et al. Estrogen-plus-progestin use and mammographic density in postmenopausal women: women's health initiative randomized trial. J Natl Cancer Inst2005;97(18):1366–1376.
- Rutter CM, Mandelson MT, Laya MB, et al. Changes in breast density associated with initiation, discontinuation, and continuing use of hormone replacement therapy. JAMA2001;285(14):1839–1840.
- Trapido EJ, Brinton LA, Schairer C, et al. Estrogen replacement therapy and benign breast disease. J Natl Cancer Inst1984;73:1101–1105.
- Ouimet-Olivia D, Van Campenhut J, Hebert G, et al. Effect of danazol on the radiographic density of breast parenchyma. J Can Assoc Radiol1981;32:159–161.
- Tobiassen T, Rasmussen T, Doberl A, et al. Danazol treatment of severely symptomatic fibrocystic breast disease and long-term follow-up: the Hjorring project. Acta Obstet Gynecol Scand1984;123(Suppl):159–176.
- Berube S, Diorio C, Masse B, et al. Vitamin D and calcium intakes from food or supplements and mammographic breast density.Cancer Epidemiol Biomarkers Prev2005;14(7): 1653–1659.
- Konez O, Goval M, Reaven RE. Can tamoxifen cause a significant mammographic density change in breast parenchyma? Clin Imaging2001;25(5):303–308.
- Tiersten A, Ng YY, Pile-Spellman E, et al. Relationship between mammographic breast density and tamoxifen in women with breast cancer. Breast J2004;10(4):313–317.
- Chow CK, Venzon D, Jones EC, et al. Effect of tamoxifen on mammographic density. Cancer Epidemiol Biomarkers Prev2000;9(9):917–921.