example_article1.txt

Abstract
Although half the world’s population will develop breasts, there is limited research documenting breast structure or motion. Understanding breast structure and motion, however, is imperative for numerous applications, such as breast reconstruction, breast modeling to better diagnose and treat breast pathologies, and designing effective sports bras. To be impactful, future breast biomechanics research needs to fill gaps in our knowledge, particularly related to breast composition and density, and to improve methods to accurately measure the complexities of three-dimensional breast motion. These methods should then be used to investigate breast biomechanics while individuals, who represent the full spectrum of women in the population, participate in a broad range of activities of daily living and recreation.

Introduction
Female breasts are specialized exocrine glands located on the anterior chest between the sternum and midaxillary line, and overlying the superficial muscles of the anterior-lateral chest wall. Although approximately half the world’s population will develop breasts, surprisingly, there has been limited scientific literature documenting the anatomical structure of female breasts or how these complex organs move. Understanding breast structure and motion, however, is imperative for numerous applications, such as reconstructing a woman’s breast following breast cancer surgery, modeling the breast to aid in diagnosing and treating breast pathologies, or simply designing effective bras that will comfortably support a woman’s breasts so she can enjoy the numerous health benefits associated with an active lifestyle. In this review, the structure and motion of the female breast are described, along with the implications for understanding how breasts affect the loading of the upper torso. Current gaps in knowledge in this field are also identified to guide future breast biomechanics research.

How Is the Breast Structured?
Female breast structure was first described by Sir Astley Paston Cooper in 1840 through his extensive anatomical breast dissections (34, 35). Although providing the foundation for our understanding of breast anatomy today, typical of the era, Cooper’s writings were purely descriptive and lacked detail about the prosections he dissected. Since 1840, only a few dissection studies have investigated the structural anatomy of the breast (9, 12, 20, 47–49, 69, 87, 110, 113, 124, 129, 131, 157, 169). Differences in the dissection techniques used in these studies, however, have led to variability in descriptions and illustrations of the internal structure and attachments of the breast. In fact, only recently has the fibro-adipose structure, fascial layers, and attachments of the breast been described in detail and supported by quantitative data and photographic dissection evidence (48, 49, 129). The main structural components of the female breast, including the external skin and the internal breast structures (fibro-adipose and fibroglandular tissue), which are all anchored to the chest wall by a complex array of anatomical attachments, are summarized below and illustrated in FIGURE 1.

FIGURE 1.
FIGURE 1.
Sagittal slice of the breast

A: illustration of a sagittal slice of the breast in line with the nipple. B: sagittal slice of an embalmed cadaveric breast in line with the nipple. The fibro-adipose tissue pockets are highlighted with dressmaker’s pins. *Cooper’s ligaments, walls of the anterior and posterior pockets.

Download figureDownload PowerPoint

Breast Skin
Covering the entire external surface of the breast is skin, which includes the nipple, areola, and general skin cover. Beneath the epidermal layer, the dermal layer of breast skin forms an interconnected mesh of collagen (primarily types I and II) and elastin fibers, which dictate the skin's mechanical behavior (76). Firmly attached over the entire surface area of the breast to the fibro-adipose tissue structure located deep to it (48), the skin covering the breast provides some anatomical breast support (48). The skin of the inframammary fold, the inferior perimeter of the breast, has an intrinsic structure known as a “zone of adherence” (12, 81, 114), which creates the characteristic curved shape of the base of the breast (114, 129). The papillary and reticular layers of the dermis of the inframammary fold contain dense, regular collagen, orientated parallel to the long axis of the inframammary fold and the skin (12, 81). This modified dermal structure is firmly adhered to the underlying muscle fascia by multiple, short fibrous connections that run between the dermis and the superficial muscle fascia (114). It functions to anchor the breast tissue to the chest wall and provides a passage for arteries and nerves to travel through on their way to the breast parenchyma and nipple areola, which limits the shearing of these structures when the breast moves (12, 114, 129).

Breast skin is thinnest in the lateral and superior breast quadrants (1.38 ± 0.24 mm) and thickest in the medial and inferior breast quadrants (1.97 ± 0.26 mm) (26). Skin thickness, however, significantly decreases with increasing age, with skin starting to thin when a woman is aged in her mid-40s (26, 154). The greatest reductions in skin thickness with increasing age occur in the lateral and medial breast quadrants (26). Thinning of skin has been attributed to reduced estrogen associated with menopause (18, 41, 60, 119). Breast skin elasticity also steadily declines with increasing age, commencing as young as the mid-20s (26). This decline in skin elasticity has been attributed to elastic fibers within the dermis degrading (14), a decline in the biosynthesis of elastin (42, 153), and reduced fibroblast tension (18, 45, 154, 155). We speculate that structural changes in dermal collagen and elastic fibers contribute to changes in breast shape, such as drooping or sagging, which are typically observed with aging (29). This notion, however, requires systematic investigation.

Understanding breast skin structure is important for applications such as designing external breast support, which is usually in the form of a bra. Decreased skin thickness and elasticity, for example, as well as increases in breast volume, found to occur with increasing age (26, 38), suggest women are likely to need increased external breast support as they age. Only limited research, however, has investigated breast support specifically for older women (73, 134, 135, 150). These studies have focused on breast support preferences of older women rather than biomechanical studies of the support requirements of these women. Given the world’s aging population and the rapidly increasing proportion of older women (167), further research in this field is warranted.

Internal Breast Structure
Deep to breast skin is a layer of subcutaneous fat (~0.5–2.5 cm thick) (50). Under the subcutaneous fat lies the “superficial fascia,” a thin fascial layer that covers the entire breast, attaching firmly to the fibrous breast perimeter around its entire circumference (49, 129). This superficial fascia is the external surface of a complex three-dimensional fibro-adipose tissue mound, which completely encases the fibroglandular tissue known as the “corpus mammae.” The corpus mammae consists of lobes of glandular tissue arranged in a radial pattern around the nipple, concentrated around the breast center (62, 67). These lobes include both lobules (milk-producing glands at the ends of the lobes) and ducts (milk passages). The ducts extend outward to the breast perimeter and widen around the nipple-areolar region to form a sac (21, 50, 56, 62). In its mature state, lobules of the corpus mammae contain 10–100 alveoli, ~0.12 mm in diameter (62). The alveoli are drained by numerous small ductiles that merge and culminate in one main duct, which dilates to form a lactiferous sinus (2–4.5 mm) that opens into the nipple (62). Primary features of the internal breast structure are illustrated in FIGURE 1.

Normal glandular tissue has been found to vary from 1 to 6.7 times the stiffness of adipose tissue (3, 5, 75, 102, 161). The values of elastic moduli reported for glandular tissue vary depending on the magnitude of strains, the loading mode (e.g., compressive versus tensile loading), and the testing method (75, 102). In vivo imaging modalities, such as breast elastography (free-hand elastography and shear-wave elastography), which measure elasticity of soft tissues, can detect differences in the mechanical properties of pathological and healthy glandular breast tissue to characterize breast lesions and assist in breast cancer diagnosis (6–8, 36, 139, 140). Finite element models of the breast tissue (137, 138) commonly use values of ~10 kPa for the elastic modulus of glandular or fibroglandular tissue, although strength properties are yet to be published.

The corpus mammae functions to synthesize, secrete, and deliver milk to a newborn child (11, 62). Its structure and function changes throughout a female’s lifespan, starting as a bud-like structure pre-puberty, which expands during puberty. The corpus mammae fully matures and remodels during pregnancy, lactation, and then again post-lactation, finally undergoing involution post-menopause (136). Involution results in regression and atrophy of the glandular tissue, up to approximately one-third of its original volume (151), with a concurrent increase in the amount of adipose tissue and reduced elasticity of the supporting connective tissue (68). The ratio of adipose to glandular tissue, however, varies among women during this period (62).

Traditionally, connective or fibrous tissue within the breast has been referred to as “Cooper’s ligaments” or “suspensory ligaments.” These “ligaments” have been described as a network of multiple scattered, cone-shaped septa, located in the deep fascia, with an apex toward the deep dermis of the breast skin (20). Recent research has revealed, however, that fibrous tissue within the breast forms a complex scaffolding, consisting of layers of fibrous tissue pockets, which are interconnected with the adipose tissue (FIGURE 2). Sharp dissection is required to remove each adipose tissue lobule from the fibrous walls of each individual fibrous pocket, suggesting the fibro-adipose pockets form a functional unit (49). Based on this new evidence of the interconnected nature of fibro-adipose and fibrous tissue within the breast, we recommend that the mechanical properties of the breast’s suspensory ligaments and adipose tissue should be investigated together rather than independently because this will improve our understanding of the integrated nature of these structures. There is a paucity of any data, however, describing the mechanical properties of breast fibro-adipose tissue (50).

FIGURE 2.
FIGURE 2.
Coronal view of an embalmed breast prosection with the skin removed, except for the nipple-areolar region

The perimeter of the breast is outlined with pins. A: the layer of the subcutaneous adipose tissue deep to the dermis. B: the fibro-adipose tissue that forms the breast mound, consisting of layers of fibro-adipose pockets. Individual adipose lobules have been removed from the walls of the fibrous pockets using sharp dissection and have been replaced by dressmaker’s pins.

Download figureDownload PowerPoint

Numerous researchers have claimed that Cooper’s ligaments provide primary anatomical breast support (e.g., Refs. 51, 82, 86, 97, 101, 142, 143). However, the mechanical properties of this connective tissue and how it interacts with the adipose tissue have not yet been reported in the literature. Further research is therefore required to quantify the material properties and mechanical behavior of both the fibrous tissue and fibro-adipose tissue units of the breast as inputs to improve breast modeling. More accurate breast models can, in turn, enhance outcomes in medicine, such as diagnosing breast pathology and surgery (3, 4, 39, 50, 67, 72, 109, 126, 128, 149, 159, 168), or in industry to improve bra design (17, 28, 57–59).

Most of the fibro-adipose pockets that form the breast mound are located either anterior or posterior to two fascial sheets, the anterior and posterior lamella, although fibro-adipose pockets are also scattered among the fibroglandular tissue (FIGURE 1) (49). The two fascial sheets, which encase the fibroglandular tissue, run in the anterior-posterior plane from the superior to the inferior breast perimeter (49, 129). These sheets are continuous with the superficial fascial system of the rest of the body, which is known as Camper’s and Scarpa’s fascia (FIGURE 1) (49). Layers of fibro-adipose pockets interconnect the anterior lamellae and the dermis, as well as the posterior lamellae and the superficial fascia of the anterior-lateral chest wall muscles. The fibro-adipose pockets located anterior to the anterior lamellae are larger but fewer in number compared with those located posterior to the posterior lamellae (FIGURE 1) (34, 35, 49). Breast size is moderately correlated to fibro-adipose pocket number (r2 = 0.7372) but only weakly correlated to fibro-adipose pocket size (r2 = −0.1758) (49). Larger breasts are therefore likely to contain a larger number of fibro-adipose pockets rather than larger fibro-adipose pockets. Based on this finding, it is suggested that the mechanism to support adipose tissue within the breast requires a specific collagen structure. It is not known, however, whether this collagen structure is due to a genetic difference in breast size or lifestyle factors (i.e., increased breast size associated with increases in BMI). Further research is therefore required to verify this notion by investigating fibro-adipose structure across the age, BMI, and breast size spectrum.

The percentage of fibroglandular tissue relative to fibro-adipose tissue within breasts varies with age, race, body mass, hormonal status, and pregnancy (13, 61, 67, 80, 156). Although highly variable, the percentage of fibroglandular tissue decreases (due to the glandular parenchyma involuting), and the percentage of fibro-adipose tissue increases as women age (13, 67, 80, 170). Breast density, therefore, decreases with age (40). For example, the average percentage of fibroglandular tissue in the breasts of younger women (n = 10, median age 31 yr, range 23–50 yr) was reported to be 36.3 ± 16.5% (78), whereas for post-menopausal women (n = 1020, mean age 59.2 yr, range 40–85 yr) this value was 19.3% (range = 13.7–25.6%) (170). These percentage values also vary according to the method used to measure tissue composition (40, 67, 78, 80, 156). That is, higher percentages of fibroglandular tissue are found using computed tomography (67, 170) and surgical dissection (156), and lower values are found using magnetic resonance imaging (80). Surgical methods are limited by their use of partial breast samples removed during breast reduction (115) or breasts that also contain pathological tissue removed during mastectomies (156). The participant numbers underlying these breast composition data sets, however, are relatively small and primarily based on post-menopausal samples.

Breast composition data have been used in modeling to assist in diagnosing and treating breast pathologies (4, 159). Breast composition data have also been used to estimate breast mass to inform the level of breast support required for bra design and breast prosthesis design (96, 101). In such studies, breast mass is typically estimated by multiplying breast volume by breast density. Breast density values, in turn, are usually based on the percentage composition of fibroglandular and fibro-adipose tissue within the breast and the relative densities of these tissues [adipose tissue: 0.5 (0.4–0.6) g/ml, mean glandular density 1.07 (1.05–1.08) g/ml] (156). Since these values are also highly variable, the absolute magnitude of breast forces determined from breast mass calculations should be viewed with caution. Furthermore, additional research is required to collect breast composition data across the age, race, and breast size spectrum to accommodate for variations in breast composition and to decrease the variation found when using different measurement techniques.

Attachments of the Breast to the Chest Wall
The breast is firmly anchored to the chest wall around the entire circumference of its perimeter (48, 87, 129). Although the perimeter structure (fascial or periosteal) and its location on the chest wall vary both regionally and anatomically (48, 129), identifying how breasts are attached to the chest wall is imperative to understand how breasts move relative to the torso. The perimeter attachments of the breast and their structure are shown in FIGURE 3. The relatively weaker fascial structure of the superior and lateral perimeter breast attachments compared with the periosteal attachment of the inferior and medial perimeter explains ptosis (drooping) of breasts that occurs with aging (29). The breast is also connected to the chest wall via loose attachments extending directly from the posterior surface of the most posterior layer of fibro-adipose pockets, which traverse between the posterior lamellae and the superficial fascia of chest wall muscles (48, 87, 129).

FIGURE 3.
FIGURE 3.
Perimeter of the breast and its attachments to the anterior-lateral chest wall

A: illustration of the perimeter of the breast and its attachments to the anterior-lateral chest wall. The dotted line outlines the pectoralis major muscle. Detailed attachments: the superior aspect of the perimeter attaches to the superficial fascia of pectoralis major and the clavicopectoral fascia. The medial aspect attaches to the periosteum of the lateral aspect of the sternum or costal cartilages of ribs 2–5. The inferior aspect attaches to the superficial muscle fascia of pectoralis major, external oblique, and rectus abdominus, and the periosteum of rib 5 or 6. The lateral aspect attaches to the superficial muscle fascia of pectoralis major, minor, serratus anterior, and external oblique muscles (48, 87, 129). B: dissection of an embalmed right breast with the breast mound (fibro-adipose and fibroglandular tissue) dissected from within the perimeter of the breast using sharp dissection. The sternal head of pectoralis major muscle has been removed from within the perimeter of the breast to expose the pectoralis minor muscle deep to it.

Download figureDownload PowerPoint

Cooper (34, 35) described the breast as attaching to the chest wall only by its posterior surface, which he stated allowed the breast to “swing from the chest wall.” However, the firm attachments to the chest around the entire breast perimeter, together with the posterior surface attachments, imply that the perimeter of the breast moves very little relative to the chest wall. Based on recent breast dissection research, we speculate that what often appears as relative movement between the breast and the chest wall, especially when women change their posture, is instead the combined effect of both the breast and the pectoralis major muscle deforming (49). Further research is required to verify this notion, particularly due to its relevance to breast modeling for diagnosing breast pathology, and understanding breast motion.

How Does the Breast Move?
Breast motion was first documented in the scientific literature in the 1970s. Dr Christine Haycock used 16-mm high-speed film to monitor the breast motion of five females, who walked and ran on a treadmill, which was set at a 1% incline (64). This research was initiated in response to an increase in females participating in sports, where exercise-induced breast pain was identified as potentially problematic (52, 63). Breasts were found to move in a three-dimensional sinusoidal pattern relative to the torso, with increased “breast bounce” (displacement) being associated with increased breast pain (64). Numerous biomechanical studies have since confirmed this sinusoidal breast movement pattern, with greater breast displacement found with increased breast mass, vertical torso motion, and lower limb cadence (22, 86, 97, 142, 144, 164). Conversely, less breast displacement was found when women wore increased external breast support, usually provided by a sports bra (22, 51, 77, 82, 84, 89, 101, 116, 120, 133, 143, 160, 164, 166, 175, 176). Graphs of three-dimensional breast displacement during treadmill running, when women ran with and without breast support, are displayed in Scurr et al. (144).

How breasts move relative to the torso has important implications for understanding exercise-induced breast pain and the design of external breast support. The total amount of breast motion during exercise is a combination of both how much the breasts are displaced relative to the torso and the total number of breast bounces. Vertical breast displacement increases during exercises that involve more vertical torso movement (i.e., greater during jumping compared with running or walking) (15, 132). In contrast, breast displacement in the sagittal and coronal planes increases during exercises involving more upper limb movement and rotation, and/or side-flexion of the torso (i.e., greater during agility tasks compared with running or walking) (132). During unsupported (bare-breasted) treadmill running, breasts have been reported to move, relative to the torso, 4.2–9.9 cm in the vertical direction, 1.8–6.2 cm in the medial-lateral direction, and 3.0–5.9 cm in the anterior-posterior direction (142–144, 166). How much of this breast motion is deformation of the breast and muscles of the anterior chest wall, rather than movement of the breast per se, is yet to be investigated. Displacement of a woman’s left and right breasts during treadmill running is usually similar, which is likely due to the relative symmetrical movement of the torso and upper limbs during this activity (89, 101, 108).

The frequency and total number of breast bounces during running are governed by step rate, because breast motion is inherently linked to foot strike. That is, during running, the torso abruptly decelerates its vertical descent when a runner’s foot strikes the ground (57, 97, 101). Although attached to the chest wall around its perimeter, inertia of the soft tissue structures causes the breast to continue moving downward before abruptly decelerating after the torso has reached its lowest point and has begun to ascend. This torso-breast time-lag causes the inferior aspect of the breast to “slap” down against the anterior abdominal/chest wall (57, 64, 97, 101, 142, 144) and is thought to be a primary cause of exercise-induced breast pain during running (64, 101, 142, 143). During 1 h of slow running, the breasts bounce ~10,000 times (101). For this reason, women who run for sustained periods of time require a high level of breast support, irrespective of their breast size.

Numerous studies have investigated the effects of different breast support conditions on breast motion (22, 25, 51, 58, 77, 82, 84, 89, 101, 105, 116, 120, 132, 133, 143, 160, 163, 164, 166, 176, 177). These studies have consistently shown that breasts move more when a woman is not wearing any external breast support (bare-breasted) compared with wearing either low external breast support (such as a fashion bra) or high external breast support (such as a sports bra). Furthermore, breast movement is greater while women wear low external breast support compared with high external breast support. In fact, a well-designed and supportive bra can decrease vertical breast displacement by ~60% relative to wearing no external breast support (25, 143, 144). Although bra manufacturers will often base marketing claims on completely stopping the breasts from moving, there is a limit to how much breast motion can be comfortably restricted due to the high composition of adipose tissue within a breast. More importantly, research investigating vertical breast displacement during deep water running revealed water-based running was associated with less exercise-induced breast discomfort compared with treadmill running, despite participants displaying similar vertical breast displacement during the two activities. Slower downward breast velocity during deep water running (−30 m/s) compared with treadmill running (−100 m/s), rather than changes in vertical breast displacement, was associated with lower levels of exercise-induced breast discomfort (97). The optimal level of breast displacement and velocity to minimize perceptions of exercise-induced breast pain is likely to depend on age, breast size, and activity type, although this optimal level is yet to be established.

Although numerous biomechanical studies have investigated breast kinematics, these studies are limited in terms of participants, measurement methods, and exercise modalities tested. Most studies have been conducted with small cohorts (usually <20) of young, healthy, relatively lean female participants, who have small- to medium-sized breasts, while the participants are walking or running on treadmills (19, 22, 25, 46, 51, 64, 77, 82, 86, 104, 116, 120, 121, 132, 133, 142–144, 164, 165, 176–178). No published study has examined breast motion displayed by women with a mean age of >35 yr, despite the world’s aging population. Furthermore, no published research was located that compared the magnitude of breast movement displayed by older women and younger women with the same breast volume. Although the magnitude of breast movement has been reported to increase with breast size, only a few of the published studies have included women with bra sizes equivalent to a DD cup or larger (15, 89, 101, 107, 108, 160, 166). This is despite bra cup sizes ranging from A up to a size I cup (147), whereby 37% of the female population have large or hypertrophic breasts (27). Furthermore, most studies have investigated breast motion during walking or running. Less than a quarter of published studies have examined breast motion during other exercise modalities, such as aquatic-based activities (97, 106, 107), jumping (15, 25, 84, 86, 106, 116, 132, 160), and upper limb movements (160). Further research is therefore required to investigate breast motion of more women, particularly those who are older, have larger breasts, and represent the body mass index of the broader female population. This research should investigate breast motion during a greater variety of activities of daily living and recreation to guide breast support and bra design for all women. Research is also required to examine women with unique breast characteristics, such as women who have had breast surgery (breast-conserving, reconstructive, or cosmetic surgery), women who have had radiation to their breasts, and women who are pregnant or currently breastfeeding. Since these women have usually been excluded from breast biomechanics research, their specific breast biomechanics and breast support needs are currently not well established. Optimizing the design of external breast support for women is important because not being able to find a comfortable bra to wear while exercising has been identified as one of the primary barriers to physical activity in women post-breast cancer diagnosis (53, 54).

Breast motion is usually measured relative to torso motion. However, the accuracy of breast displacement data relative to the torso is questionable because researchers have typically used traditional biomechanical methods, which have been developed to quantify rigid body motion, to measure breast motion. As described above, breasts are highly complex, non-uniform, soft-tissue masses that lie over deformable muscles; are anchored to the chest wall by attachments with varying material properties; and both the breast and torso move in complex three-dimensional patterns. Most published studies report using single markers placed on the nipples (either under or over a bra) to measure breast displacement relative to a torso marker. A single reference point on the nipple is unlikely to accurately capture the complexity of breast motion. A few studies incorporated either additional markers around the nipple (5 markers) (84, 86, 178) or multiple markers on a crop top covering the breasts (54 markers) (1) to track breast motion. Although these researchers found differences in the displacement of the nipple marker compared with the other breast markers, the markers were placed on top of the bra (which can move relative to the breast) and not directly on the breast. The mass of the 54 markers (0.3 g each) tracked by Arch et al. (1) is also likely to have affected the motion of the breasts under the crop top. There are also errors inherent with marker aberration, particularly when markers used to define the torso segment are placed on areas of high adipose tissue, such as the anterior abdominal wall. Consequently, the accuracy of torso and breast displacement data, and any variables derived from these data such as velocity and acceleration, must be considered with caution. These data are unlikely to be accurate in women with larger, more deformable breasts (older women with large ptotic breasts) or those with more abdominal adiposity. Researchers have attempted to directly measure breast acceleration using accelerometers, but this has also been limited by the soft, deformable nature of breasts (65, 111). Further research is therefore required to develop reliable and valid methods, which can account for the highly complex, non-rigid anatomy of the breast and its attachments to the torso, to more accurately measure three-dimensional torso and breast motion. Measurements techniques such as speckle tracking photography (2, 85) could be used for this purpose. Any such method needs to be used to quantify breast kinematics, particularly of older women and those with larger breasts and higher adipose tissue, while these women perform a variety of activities of daily living and recreation. Accurate measurement of breast kinematics is also essential as input to determine how breast motion influences the loading of the upper torso.

How Do Breasts Affect the Loading of the Upper Torso?
The forces generated by breasts have been calculated by researchers using Newtonian equations of motion (force = mass × acceleration) together with estimates of breast mass and breast acceleration (50, 57, 101). Knowledge of breast anatomy to estimate breast mass, as well as breast kinematic data (i.e., acceleration), is therefore essential to understand how the breasts can affect loading of the torso and, in turn, upper torso musculoskeletal structure and function (30, 32, 33, 58, 95, 96, 101).

Breast Mass
Breast mass can increase with increases in body weight (increased BMI) (27) and with the increased breast volume that occurs during pregnancy and breast-feeding, secondary to increases in glandular tissue and milk production (62). Breast mass can also be affected by the increase in breast volume that can occur during the menstrual cycle (41, 127), as well as by the increase in breast density that can occur with breast cancer (13, 40, 168). As stated previously, breast mass is typically estimated by multiplying breast volume by breast density (50, 57, 101). Breast volume has been calculated using a variety of methods (24). Examples of these methods include water displacement (16, 91, 148); anthropometric measurements (146, 148, 158, 162, 174); medical imagining techniques such as computerized tomography scans (67), mammograms (70), and magnetic resonance imaging (130, 173); and three-dimensional scanning of the external surface of the breasts (27, 28, 66, 74, 79, 83, 94, 95, 98, 103, 112, 118, 123, 128, 145, 152, 171, 172). Three-dimensional scanning of the external surface of the breasts is currently the most common method used to calculate breast volume due to its geometrical accuracy and ease of use (FIGURE 4) (27, 28, 74, 79, 83, 94, 95, 98, 172). There are limitations, however, in the accuracy when women with large ptotic breasts are scanned, particularly when the lower aspect of the breast sits on the woman’s anterior abdominal wall. This blocks a scanner’s field of view of the inferior aspect of the breast, thereby leading to underestimates of breast volume (28, 37, 83, 94, 98, 152, 171). There is also debate regarding how the posterior wall of a scanned breast image should be shaped when forming a breast model to calculate breast volume and how this is affected by the position of a participant while she is being scanned (i.e., prone versus standing) (28, 94, 98, 172). The effects of different scanning methods and participant postures on the accuracy of measurements derived from three-dimensional scanning of the breasts and torsos of women are described in detail elsewhere (94).

FIGURE 4.
FIGURE 4.
Three-dimensional scanning of the external surface of the breasts

A: example of a participant standing on a turntable while her breasts and torso are scanned using a handheld scanner. B: a three-dimensional scan with the breast perimeter outlined in red. C: the three-dimensional model of the breast from which breast volume is calculated.

Download figureDownload PowerPoint

Breast volume, calculated from three-dimensional scans of the breasts of 487 Australian women (age range 18–80 yr; BMI range 8–42 kg/m2), was found to range from 48 to 3,100 ml/breast (27). There was no significant difference in the mean breast volume between the women’s left (653 ml, range 70–2789 ml) and right (647 ml, range 48–3100 ml) breasts. Based on this data set, Coltman et al. (27) classified the size of women breasts as small (breast volume of <350 ml; ~28% of participants), medium (350–700 ml; 37% of participants), large (700–1,200 ml; 24% of participants), and extra-large (hypertrophic; >1,200 ml; 11% of participants). The data set was also graphed to illustrate the distribution of breast volume across the age spectrum (27). Although breast volume data are relatively accurate, as stated above, breast density is highly variable, making reliable and accurate estimates of breast mass difficult. Breast mass can be directly measured by using mastectomy specimens (172), although such tissue is likely to contain pathological tissue. Furthermore, directly measuring breast mass is not an option when healthy women are investigated. More accurate methods to determine breast mass are therefore required so we can better estimate the loading of the upper torso caused by variations in women’s breasts.

Loading of the Upper Torso
Irrespective of difficulties in estimating breast mass, when a woman stands upright, a flexion torque is generated about her thoracic spine due to gravity acting through the center of mass of her breasts (30, 32, 95, 101). Given that torque depends on a force (breast weight) multiplied by a moment arm (perpendicular distance from the breast center of mass to the thoracic spine), the flexion torque about the thoracic spine is greater in women with large breasts (higher breast weight and longer moment arm) compared with women with small breasts (less breast weight and shorter moment arm) (30, 32, 44, 101, 141). Excessive loading of the muscles and soft tissue structures of the thoracic spine caused by this thoracic flexion torque has been associated with secondary changes in posture and compromised movement of the vertebral column and upper limbs in women with large breasts (30, 32, 95, 150). In fact, women with large and hypertrophic breasts report experiencing more upper torso musculoskeletal pain and display increased thoracic kyphosis, as well as decreased shoulder range of motion and scapular retractor endurance strength, compared with their counterparts with smaller breasts (30, 32, 95, 141). Increases in breast mass during pregnancy and breast-feeding can also increase loading on the thoracic spine and have been associated with increased thoracic kyphosis angle and musculoskeletal pain (122, 125).

These changes in thoracic posture, compromised shoulder range of motion, and reduced scapular retractor endurance strength displayed by women with large breasts have provided evidence upon which to base strategies to treat symptoms of upper torso musculoskeletal pain in these women. For example, the thoracic flexion torque, and in turn loading on the thoracic spine, can be reduced by decreasing breast mass through weight loss or breast-reduction surgery (10, 30, 32, 43, 44, 71, 95, 117). Alternatively, ensuring women with large breasts wear highly supportive bras to bear some of the breast weight can also reduce the thoracic flexion torque (30, 32, 95, 101). Unfortunately, women are often found to be wearing poorly fitted or unsupportive bras (31, 55, 88–90, 100). Evidence-based education resources are therefore needed to guide women on the importance of correct breast support and bra fit (23, 92, 93, 99) to reduce unnecessary loading of the upper torso.

Summary
Biomechanical research on the structure and motion of the female breasts, and how this structure and function affects the loading of the upper torso, has made substantial contributions to medicine, health, and the apparel industry. Knowledge of breast structure, composition, density, and deformation have provided evidence for breast surgery, as well as for breast models that can enhance the diagnosis and treatment of breast pathology. Research on breast motion has provided scientific evidence upon which to design sports bras to improve external breast support for women so they can exercise in comfort. Understanding the loading of the upper torso caused by breast weight has also provided evidence upon which to develop strategies to treat breast-related musculoskeletal issues in women with large breasts. Despite these advances, there are still large gaps in our knowledge pertaining to breast structure, particularly gaps in our knowledge related to breast composition and density. There is also an urgent need to improve biomechanical research methods that can accurately measure the complexities of three-dimensional breast motion, particularly while women participate in a broad range of activities of daily living and recreation. These studies should recruit participants who represent the full spectrum of women in the population, including women who are older, have larger breasts, and have a higher body mass index, as well as those who have unique breast characteristics (e.g., following breast surgery). Further research is encouraged to fill these knowledge gaps and improve breast biomechanics research methods to ensure accurate, valid, and reliable data pertaining to breast structure, motion, and loading, which can be used to enhance the breast health of all women, irrespective of age, breast size, health status, and physical activity pursuits.

No conflicts of interest, financial or otherwise, are declared by the author(s).

D.E.M. and J.R.S. prepared figures; D.E.M. and J.R.S. drafted manuscript; D.E.M. and J.R.S. edited and revised manuscript; D.E.M. and J.R.S. approved final version of manuscript.