Newly Identified Gross Human Anatomy: Eight Paired Vestigial Breast Mounds Run along the Embryological Mammary Ridges in Lean Adults.

INTRODUCTION

If you can measure it, measure it. If you can’t measure it, measure it.

—Nobel Laureate Surgeon, Charles B. Huggins, MD1
Descriptions of accessory breast tissues appear in historic and current literature. The majority pertains to women, but rare reports describe male findings.2 Although some publications associate anomalous breast tissues with systemic conditions, the overwhelming majority use a variety of terms to describe benign vestiges of normal breast tissues (eg, ducts, fatty mounds, or accessory nipples).3,4 Data on the incidence of accessory breasts are lacking. However, in a 1907 study, Japanese researchers reported that 5.19% of women and 1.68% of men presented with “polymastia.”5 A 1924 report showed that “polymastia” was a multigenerational phenomenon. In 1939, one author found a 1%–2% prevalence of “supernumerary breasts” among Whites.6,7 The presence of “ectopic nipples” or “polythelia” has been documented in both genders, and “supernumerary” structures have been reported occurring along the embryological milk line.8–11 Several well-known anatomic atlases have presented illustrations that visually summarize accessory breasts along curving “milk lines” descending from axillae to proximal thighs bilaterally, including a now-classic diagram by Frank Netter.12 By contrast, no significant discussion or summation of findings is available for men. Other mammalian species exhibit up to nine pairs of female mammary glands and male counterparts usually develop similar nonlactating structures.13
To date, the literature has described human accessory breast tissues as “anomalous” or “pathologic.” However, given commonalities with other mammals and informed by clinical findings, we postulated that remnants of breasts in the form of paired adipose mounds along the mammary ridges could represent previously unrecognized normal anatomy in all humans. Unfortunately, we could find no system for measuring relative subcutaneous fullness in a quantifiable, reproducible way to analyze adiposity for the present study. Inspired by Dr. Charles Huggins, we recognized the need to devise a new system of measurement.1
Although current culture is obsessed with breasts, no author has presented an overarching, embryologically-based theory to explain the array of anatomic findings present in the general population or changes in breast and body shape that occur over time. Such insights could provide a foundational understanding of basic anatomy, potentially improving clinical analysis and management of a variety of conditions.
While caring for patients with HIV lipodystrophy, the principal author (DT) recognized that focal paired adipose mounds became hypertrophic along the embryonic milk line concurrent with wasting of the surrounding fat. Author DT published an illustration of the mammary ridge concept and postulated that the mounds were embryologically similar, of breast origin, and resistant to HIV-atrophy.14 Concurrently, HIV-negative men presenting for elective liposuction almost always exhibited paired mounds of adipose in the same locations seen in virus-induced lipohypertrophy. However, unlike in chronic HIV, these mounds were diminutive and grew slowly. In both cases, these paired protuberances occurred at consistent anatomic locations descending from axillae to anteromedial thighs, with the distinct appearance of teats seen in other placental mammals.
We hypothesized that a consistent pattern of small vestigial-breast fat mounds is present in lean healthy individuals and represents previously unrecognized normal human anatomy. To test the premise, we performed a quantitative anatomic analysis using topographic grid mapping, pinch testing of subcutaneous fullness, and standardized photography of the anterior torsos of male and female study participants.

MATERIALS AND METHODS

Study Population
Healthy medical models (10 men, 10 women) ranging in age from 18 to 33 years were recruited from a well-established online modeling website (www.modelmayhem.com) based upon their declared comfort with being photographed nude. The study set was intentionally restricted to young, lean, nulliparous individuals of European extraction, to minimize variation regarding age, BMI, hormonal influence, race, and ethnicity.
This prospective study of human anatomy in nonpatient, healthy, paid, medical models meets standards outlined in the World Medical Association’s Declaration of Helsinki – Ethical Principles for Medical Research Involving Human Participants. The senior author (DT) is CITI-certified for conducting a biomedical research investigation using human participants and attests to meeting appropriate ethical guidelines. All participants were fully informed of the study design and protocol, gave permission for data analysis and publication of de-identified imagery, and received moderate compensation.

Inclusion and Exclusion Criteria
We measured height, weight, and fat content for each potential participant and excluded individuals if fat content was greater than 14.0% for men or 23.0% for women. Participants were required to be postpubertal and free of known systemic disease, chronic infection, endocrine aberration, or visible skeletal and skin abnormalities. We also eliminated individuals with anterior torso tattoos, piercings, surgical scars, other trauma to the body surface, or steroid exposure. One registered man was replaced as a participant when he revealed receiving anabolic steroids in high school. We recruited women who had never been pregnant beyond 8 weeks from conception.

Anatomic Documentation
To start, we captured a baseline sequence of faceless rotational photographs of the circumferential torso for each participant, using strictly standardized protocols developed by our laboratory and published over the last decade.15–17

Anatomic Grid Mapping
The primary author (DT) used Sharpie marking pens, a bubble-level ruler with millimeter increments, and tape measures to ink a grid onto the anterior surface of each participant, anchored by specific anatomic features. To establish the lateral border on each side, we drew a black line from the distal end of the clavicle, wrapping across the insertion of the pectoralis major up and into the apex of the axilla. DT then used single-point perspective to extend the mid-lateral line of the thorax and thigh to the level of the superior border of the patella. To establish vertical sub-divisions, each of the five major horizontal lines was measured from the lateral line to the midline and divided into five equal segments on each side using blue hatch marks. We connected these breakpoints with blue pen to create five vertical columns on each side of the midline. We labeled columns of cells 0–9, starting at the participant’s right acromion, then labeled rows A through Y in cephalocaudal progression. Finally, during isometric contraction of the rectus abdominus, DT outlined the perimeters of each segment in red (if palpable). The resulting anterior grid map contained 250 rectilinear cells (Fig. 1). (See appendix 1, Supplemental Digital Content 1, which displays a stepwise checklist of tasks to ensure that each participant was mapped and photographed consistently. http://links.lww.com/PRSGO/B802.)
We captured a set of standardized rotational photographs to document the grid on each participant. Although quadrilateral cells varied widely in their dimensions among participants (resulting from variations in height and bodily proportions), anatomic features appeared in consistent cells on the grid map (eg, male nipples appeared in cells E2 and E7, whereas female nipples were invariably located within 2 cm of the intersection of rows D/E with columns 2/3 and 6/7).

Data Collection from Each Grid Cell
We collected data with participants standing, as this is the gravitational orientation used socially and medically to evaluate and compare body shapes. We employed a standardized data collection sheet and first evaluated each grid cell for the number of pigmented nevi (potential forme fruste accessory nipples), then counted shafts of hair. (See appendix 2, Supplemental Digital Content 2, which displays the standardized data collection sheet. http://links.lww.com/PRSGO/B803.) We then used anthropometric calipers (Harpenden Skinfold Caliper, John Bull, British Indicators, Ltd., West Sussex, UK) capable of reading to 0.2 mm to pinch-test subcutaneous fullness (Fig. 2).
Some cells exhibited significant skin hyperelasticity, with tissues that protruded significantly beyond the pincer platforms during compression. When this occurred, we recorded two separate pinch-values to better capture relevant data. The first measurement was made with protrusion of tissues. Then, a modified “containment pinch test” was performed by applying just enough pressure to prevent protrusion and keep the full volume of skin and fat between the pincer pads, yielding a higher pinch thickness (Fig. 3). For consistency, DT performed all measurements for the study, with a second member of the team witnessing the caliper values as they were called out to scribes who entered the data into alphanumeric spreadsheets matching the grid maps on participant’s bodies. We used four separate spreadsheets to record vertical pinch thickness (from both standard and modified containment techniques), horizontal thickness (both standard and modified values), number of visible hair shafts, and the number and characteristics of any moles. We videotaped each session.

Topographic Mapping of Focal Fat Mounds and Final Documentation
Once we obtained data for each participant, DT topographically plotted the location and thickness of the focal fat mounds onto the grid-marked skin surface using contour mapping techniques published recently by our group.18,19 The boundaries and projections of each fat pad were documented on the skin surface in their actual locations using contour rings at 1-cm increments of fullness above baseline (Fig. 4). Finally, we documented the topographic maps with additional sequences of standardized photographs (Fig. 5).

Statistical Methods
Review of existing anatomical and surgical literature, and Google image-searches for “mammary ridge” and “milk line” (performed 12 July 2020) allowed us to map the location of the mammary ridge against the collected data. Using Photoshop, we standardized and summed the 10 highest-ranked, published illustrations to match the proportions of study participants. Figure 6A shows this gridded overlay. Ninety-two of 250 cells had more than 50% of their area filled by the overlay diagram, were plotted in black (Fig. 6B), and identified as the predicted paths for adult mammary ridges. To test our hypothesis that the anterior torso’s pinch-tested mounds relate to the embryological mammary ridge, we used linear regression to evaluate the relation between pinch thickness in each cell and the predicted locations.20–23 (See appendix 3, Supplemental Digital Content 3, which displays the statistical methods. http://links.lww.com/PRSGO/B804.)

RESULTS

Study Population, Fat Pad Locations, and Subcutaneous Fullness
Age, height, weight, fat content, and BMI were determined and tabulated for both male and female participants. (See table 1, Supplemental Digital Content 4, which displays the anthropometric data obtained from the 10 lean male study subjects. http://links.lww.com/PRSGO/B805.) (See table 2, Supplemental Digital Content 5, which displays the anthropometric data obtained from the 10 lean female study subjects. http://links.lww.com/PRSGO/B835.)
Pinch testing revealed paired adipose mounds in the following loci in men: axilla (100%), lateral chest wall tail (20%), sub-areola (100%), anterior-inferior chest wall (100%), upper abdomen (100%), lower abdomen (100%), pubis (100%), and anteromedial thigh (100%). Every woman participating had focal fullness in 100% of these same locations. Importantly, no focal subcutaneous fullness was found anywhere else on the anterior surface of the body in any participant. Therefore, eight specific discreet anatomic loci accounted for 100% of focal adipose among participants. Mounds were always in a curvilinear configuration, extending from axillae to the anteromedial thighs bilaterally (Fig. 7). We saw no relationship between body fat content and the presence, number, or size of the mounds. Tables 1 and 2 display average pinch data from men and women participants, respectively.

Statistical Findings
Table 3 displays the estimates from the separate OLS multiple regression models. In men, the difference between the average pinch measurement for mammary ridge cells and nonridge cells was significant when the covariate identifying the augmented measurements was included (0.027) and when it was not included (0.023). In contrast, among women, the difference between the average pinch measurements for mammary ridge cells and nonridge cells was not significant when the covariate was included in the model (0.115); however, it was significant when the covariate was not included (<0.001). One possible explanation for this may be the greater proportion of female measurements with pinch values of zero, where the skin was directly attached to the underlying fascia, when compared with the male measurements; 52.0% of female horizontal and vertical measurements had pinch values of zero, compared with only 18.7% among men.

Hairs, Moles, and Hyperelasticity
Counts of hairs and moles in cells along the mammary ridge appeared greater than in nonridge cells, but our limited sample size prevents us from drawing distinct conclusions. However, those moles present atop focal mounds often had the cobbled appearance of nipples (Fig. 8). We also determined that hyperelastic skin was consistently identified along the mammary ridge (Fig. 9). Supporting our findings, massive weight loss patients can exhibit pendulous focal fullness, laxity, and stretch marks running along the curved mammary ridge lines in the same pattern as our lean participants (Fig. 10).

DISCUSSION
We describe a new method of determining and mapping human subcutaneous pinch thickness by plotting a standardized two-dimensional grid directly onto the entire anterior skin surface of torsos. Pinch testing confirmed that a consistent and previously unrecognized pattern of focal fat mounds is present in our participants, regardless of gender. Paired mounds occurred at regular intervals in predictable locations on the long-known paths of the embryological mammary ridges on the anterior torso bilaterally. This suggests that vestiges of primordial breast fat remain present and anatomically relevant in adults, not having undergone involution during fetal development as previously presumed. Also supporting our findings and hypothesis, review of the English language literature revealed that all reports of accessory breasts on the anterior torso occurred in one of these same eight locations listed and seen in Figure 9. Finally, the concomitant findings of increased hair counts, increased presence of pigmented nevi, and skin hyperelasticity along the same paths provide strong supporting evidence for diminutive functionless breast mounds.
Rather than think of “accessory” breasts as “extras” or “anomalies,” we suggest that chains of vestigial breast mounds of variable sizes represent previously unrecognized normal gross human anatomy. Until now, their consistent presence in healthy adults appears to have gone unnoticed by the scientific community for at least two reasons. The distinction between mounds cannot easily be pinch-identified in any but the leanest individuals. Also, the complex three-dimensional undulating surface of underlying bones and muscles makes it difficult to see the pattern of these small protrusions running linearly down the torso.
Our observations align with previous publications regarding the embryonic mammary ridge, including a developmental anatomy textbook from 1948, the fourth edition of Principles of Surgery from 1984, and a more recent report of an ectopic breast from 2012.23–25 However, these only invoke descriptions of the mammary ridge to explain abnormalities, not as rationale for normal fat pad anatomy. Importantly, we agree with at least one other author that focal accessory breast mounds should not be considered pathologic and that these mounds do not warrant “medically necessary” removal, except in rare cases when diagnostic tests suggest otherwise. Unlike primary breasts, mammary ridge mounds do not normally include ductal elements and have a negligible risk of malignant transformation.26
We also suggest that vestigial mounds along the milk line, regardless of gender, should be less responsive to weight loss because we found fat mounds in our leanest participants, agreeing with a study published in 1980, which showed little association between body fat content and breast size.27 Also, in HIV lipodystrophy, body-wide subcutaneous fat can be lost entirely, while focal mounds along the mammary ridge continue to enlarge in an aberrant breast-like fashion, unhampered by disease progression.14 Clearly, vulnerabilities of these two subpopulations are vastly different. Previously published work, our pinch data, and clinical observations all support the concept that there are likely two major subpopulations of subcutaneous yellow fat: diffuse, body-wide subsurface adipose and paired focal mounds of primordial breast fat. Perhaps adipose should now be classified into four major subtypes: brown fat of newborns, internal visceral fat, focal yellow embryonic breast fat, and diffuse yellow subcutaneous fat.
Our observations also have implications for donor site selection for harvesting fat grafts. Mammary ridge fat and visceral fat appear to survive both disease and autologous transfer procedures better than general subcutaneous fat,14,28 with the exception that in our experience, fat harvested from the lower abdominal mounds survives less-well for unknown reasons.
Finally, we propose that variability in relative size of mammary ridge mounds accounts for the wide variety of body shapes seen throughout society. Factors like familial patterns of focal fat fullness, hormonal shifts like pregnancy, andropause, and menopause, some disease processes like diabetes and Cushing’s disease, and several pharmaceutical exposures can all lead to body shape dissatisfaction and become chief complaints for those seeking surgery. We also have a strong suspicion that perception of sexuality (“gaydar”) may be primarily determined by subconscious recognition of patterns of relative fullness (or lack of fullness) of specific focal mounds, though further work is needed to draw conclusions. Based upon these preliminary observations, we postulate that embryonic fat pad differentiation may be temporally related to differentiation of the hypothalamus and suggest this link be investigated.

Limitations of the Study and Future Investigation
The current study is limited in several ways. Our sample size is relatively small, but among our 20 qualifying subjects, we gathered more than 20,000 unique data values for analysis, and the consistent pattern of focal adiposity shows great statistical significance. Participants were not selected at random by an individual blind to the purpose of the study; however, after meeting the stringent inclusion/exclusion criteria, were enrolled sequentially as they responded to our recruitment postings, without any additional selection from our team.
Although it may be perceived that having a single investigator performing all measurements could induce bias, we anticipated this concern and had every pinch measurement witnessed by another member of the team. We also video documented every skin evaluation made and each pinch-test performed, so that values from every cell can be validated.
The focus on participants of ethnic European extraction restricts our ability to extrapolate findings to other populations. However, the senior author has noted that findings among the diverse patients in his surgical practice consistently match this study’s findings.
We conducted pinch testing only on the anterior surface of the body for the present study, but clinical experience (especially with patients having HIV lipodystrophy) reveals similar paired adipose mounds on the posterior body surface. Our group has therefore begun to define fat pad anatomy for the remainder of the torso and all four limbs.

CONCLUSIONS
Grid mapping and pinch testing show that healthy lean adults consistently exhibit eight pairs of vestigial breast mounds along the previously described embryological milk line. These focal areas are associated with increased overlying hair counts and the presence of moles that may represent forme fruste areolas or nipples. Also, skin along the milk line is hyperelastic, perhaps providing less of an impediment to volumetric expansion of breasts during puberty and pregnancy. Variable fullness of these paired mounds likely accounts for the full spectrum of human body shapes, including anatomic differences that allow us to perceive genders, states of health or disease, relative ages, and the sexualities (in our strong, albeit preliminary observations) of those around us. We believe that our findings provide a framework to better understand adipose anatomy and physiology within a newly envisioned embryological context that honors all previously published observations we have found.
With the recent burst of interest in liposuction and autologous fat transfer, these anatomic findings provide insights to improve surgical planning and three-dimensional outcomes. Finally, whether for reconstructive or aesthetic needs, selection of fat grafts from donor sites with specific adipose physiologies should enhance graft stability and long-term results.

ACKNOWLEDGEMENTS
We thank the American Institute of Bisexuality of Los Angeles, California for critical financial support for the development of the alphanumeric grid mapping technique, for hiring of medical models, for data collection and analysis, and for preparation of this article. John Sylla of Los Angeles has provided important ongoing guidance and insights. The Marena Group of Atlanta, Georgia provided support for project development and imaging refinements. Early co-author Louis Keith, MD, PhD, ScD (Hon) of Chicago helped with project design, article preparation, and cross-checking of data, but sadly died before project completion and article finalization. The authors also thank the senior faculty of The Kinsey Institute for guidance and encouragement, including Julia Heiman, PhD and Stephanie Sanders, PhD, as well as Erick Janssen, PhD, who has since relocated to the University of Leuven. Virginia J. Vitzthum, PhD of the Department of Anthropology at Indiana University provided the calipers used for data collection and gave important insights regarding pinch testing. The anatomic analysis techniques presented here were created as part of a larger fully funded initiative to formalize the documentation and quantification of living human anatomy, and to find associations between anatomy, behavior, gender, and sexuality. The present study is part of that project, and conducted in collaboration with The Kinsey Institute of Bloomington, Ind., with major general administrative support from the American Institute of Bisexuality of Los Angeles, California. The 803 Foundation also received unrestricted funding from the Marena Group of Atlanta, Georgia, which made preparation of this article possible. We are grateful to Emmett Cooney and Family for unrestricted financial support that was critical for article completion and preparation of illustrations. No individuals from these funding sources had any role in background research, project direction, or preparation of the final article and its content.

Supplementary Material