Cosmetic

Nonsurgical Breast Enlargement Using an

External Soft-Tissue Expansion System
Roger K. Khouri, M.D., Ingrid Schlenz, M.D., Brian J. Murphy, M.D., and Thomas J. Baker, M.D.
Key Biscayne and Miami, Fla.

Less than 1 percent of the women interested in having Approximately 16 to 19 million women in
larger breasts elect to have surgical augmentation mam- the United States between the ages of 18 and
maplasty with insertion of breast implants. The purpose of
this report is to describe and test the efficacy of a non- 49 have an expressed interest in breast enlarge-
surgical method for breast enlargement that is based on ment1,2; however, despite a resurgence in pop-
the ability of tissues to grow when subjected to controlled ularity, only about 130,000 (0.7 percent) of
distractive mechanical forces. Seventeen healthy women these women underwent surgical breast aug-
(aged 18 to 40 years) who were motivated to achieve breast mentation in 1998.3 Reluctance to undergo
enlargement were enrolled in a single-group study. The
participants were asked to wear a brassiere-like system that surgery for cosmetic reasons, perceived adverse
applies a 20-mmHg vacuum distraction force to each sequelae from the implants, and cost are the
breast for 10 to 12 hours/day over a 10-week period. Breast most cited deterrents to this surgical recourse.2
size was measured by three separate methods at regular We developed a system and a method of
intervals during and after treatment. Breast tissue water
external soft-tissue expansion and tested its ef-
density and architecture were visualized before and after
treatment by magnetic resonance imaging scans obtained ficacy as a nonsurgical alternative for breast
in the same phase of the menstrual cycle. Twelve subjects enlargement. The principle behind this ap-
completed the study; five withdrawals occurred due to proach is the capacity of tissues to grow when
protocol noncompliance. Breast size increased in all subjected to sustained, low-level, mechanical
women over the 10-week treatment course and peaked at
week 10 (final treatment); the average increase per
distraction. For centuries, tribes from several
woman was 98 ⫾ 67 percent over starting size. Partial cultures have applied this principle to enlarge
recoil was seen in the first week after terminating treat- various body parts.4,5 Surgically implanted tis-
ment, with no significant further size reduction after up sue expanders are now routinely used in plastic
to 30 weeks of follow-up. The stable long-term increase in surgery to incrementally increase the amount
breast size was 55 percent (range, 15 to 115 percent).
Magnetic resonance images showed no edema and con-
of skin and soft tissue available to perform
firmed the proportionate enlargement of both adipose multiple staged reconstructive procedures.6 – 8
and fibroglandular tissue components. A statistically sig- Orthopedic experience with the Ilizarov proce-
nificant decrease in body weight occurred during the dure has demonstrated the feasibility of length-
course of the study, and scores on the self-esteem ques- ening extremities by a process of gradual dis-
tionnaire improved significantly. All participants were
very pleased with the outcome and reported that the de-
traction that grows the bones and associated
vice was comfortable to wear. No adverse events were soft tissues.9 –11 New devices have extended the
recorded during the use of the device or after treatment. use of this principle to advance facial bones
We conclude that true breast enlargement can be and correct retruded faces.12,13 Cell biologists
achieved with the daily use of an appropriately designed have devoted considerable research toward
external expansion system. This nonsurgical and nonin-
vasive alternative for breast enlargement is effective and elucidating the mechanism of mechanotrans-
well tolerated. (Plast. Reconstr. Surg. 105: 2500, 2000.) duction, the process by which mechanical ten-

Received for publication December 1, 1999; revised February 7, 2000.

Presented in part at the Annual Meeting of the American Society for Aesthetic Plastic Surgery, in Dallas, Texas, May 16, 1999, and at the 10th
Anniversary Meeting of the European Association of Plastic Surgeons, in Madrid, Spain, May 21, 1999.
Roger K. Khouri, M.D., is a shareholder of Biomecanica, Inc. None of the other authors have any financial interest in or have received any
direct benefit from Biomecanica, Inc., except for Ingrid Schlenz, M.D., who received partial salary support from Biomecanica, Inc., during the
study period.
2500
Vol. 105, No. 7 / NONSURGICAL BREAST ENLARGEMENT 2501
sion is converted into growth-promoting sig- of 15 to 25 mmHg. To document protocol
nals.14 –17 compliance, the microcomputer records tem-
Current therapeutic applications of tension- perature and pressure every 10 minutes and
induced tissue growth require surgical inter- stores the data in its memory.
vention to insert either an inflatable silicone A number of suction-based devices have
shell as the force-transducing device or bone touted the ability to induce breast enlarge-
pins and screws as links to the distraction ment.18 These are considered to be mostly in-
frame; however, the approach presented here effective novelty items and are ridiculed by the
is entirely nonsurgical. Sustained, low-level establishment in plastic surgery.19 Although
negative pressure (vacuum) provides the out- the basic idea behind stretching to induce en-
ward distractive force that stimulates breast tis- largement is sound, these novelty devices lack
sue enlargement. The nonsurgical breast- the necessary design elements to appropriately
enlargement system was designed by one of the deliver, over a prolonged period of time, a
authors (R.K.K.); in a pilot study on two sub- sustained, controlled mechanical distraction to
jects (four breasts), he found evidence that the cause breast growth.
device was effective (unpublished data). The We built the system by deliberately taking
present study represents an independent test into account known physiologic constraints
of its safety and efficacy in a larger population. and bioengineering principles. Pilot testing of
a succession of prototypes led us to identify
MATERIALS AND METHODS and refine four novel critical design features
that ensure safe and effective function (U.S.
Soft-Tissue Expansion patent #5536,233; U.S. patent #5662,583; U.S.
The nonsurgical breast-enlargement system patent #5676,634; U.S. patent #5695,445; U.S.
incorporates two semirigid plastic domes, each patent #5701,917; and other patents pending).
slightly larger in volume than the correspond-
ing breast to be enlarged. A brassiere garment Balancing of Forces and Optimization of Pressures
supports the two domes, and the device is worn The total outward force (Fout) exerted by the
like a brassiere (Fig. 1). Each dome has an vacuum on the breast is equal to the product of
outlet port for vacuum application and a gel- the aperture area of the dome (A) and the
filled bladder at the rim that circumscribes the vacuum pressure (Pv). This force is balanced by
outer margins of the breast. An adhesive, hy- an equal counter-force against the chest skin
poallergenic silicone gel applied to the skin- under the bladder rim (Fin) that is equal to the
contact surface of each bladder maintains a product of the area of the bladder rim (R) and
vacuum seal with the skin. A battery-powered, the pressure transferred to the skin (Ps). Cap-
microcomputer-controlled vacuum pump is illary perfusion starts to drop precipitously
connected to outlet ports on each dome by when the pressure (whether positive or nega-
plastic tubing. Pressure sensors and relief tive) exerted on the skin and underlying tissues
valves are used to maintain a vacuum pressure exceeds 20 to 30 mmHg.20 –23 Sustained pres-

FIG. 1. (Left) Diagram showing the translucent plastic domes and gel-filled bladder rims
placed over the breasts. The microcomputer-controlled vacuum pump fits inside a pocket of the
brassiere garment. (Right) Photograph of the nonsurgical breast-enlargement system as worn by
a participant in the study.
2502 PLASTIC AND RECONSTRUCTIVE SURGERY, June 2000
sures above this limit lead to tissue damage. To Control of Shear Forces on the Skin
maximize distraction without compromising The inward pull of expansion stretches the
the circulation of the distracted breast or that breast skin close to the limit of its elastic defor-
of the compressed skin under the rim, the mation.
absolute values of Ps and Pv must equal this At the periphery of the breast, if the skin is
upper limit of tissue tolerance. This optimiza- held firmly by the inner lip of the bladder rim,
tion of the absolute pressures requires the area this stretch imparts a strong shearing force (Fs)
of the bladder rim (R) to be equal to the to the skin, which can combine locally with the
aperture area of the dome (A), as shown in the counter-force (Fin) to cause peripheral skin
following equations. blistering and breakdown.24 To reduce this po-
tentially damaging shear force, the peripheral
F in ⫽ Fout breast skin under the inner lip of the rim
should not be fixed but allowed to move in-
A ⫻ P v ⫽ R ⫻ Ps ward. To accomplish that, the rim must deflect
radially inward for the distance needed to re-
If P v ⫽ Ps, then R ⫽ A cruit additional skin and reduce the shear
stress to a tolerable level. In addition, this fea-
The skin contact area of the rim bladder is ture reduces skin stretching and the resultant
therefore designed to be approximately equal undesirable skin expansion, while focusing the
to the area of the aperture of the dome (Fig. distraction on the deeper breast tissue.
2). This constraint can only be avoided if the Through a succession of prototypes, we found
pressure is allowed to alternate, decreasing the that a rim bladder approximately 2.5 cm high
vacuum level every few minutes to allow the provides the necessary arc of inward deflection
skin under the rim to reperfuse, and then re- (Fig. 3).
establishing the vacuum for the next cycle. We
elected not to use alternating pressure because Pressure Distribution and Avoidance of
of the power drain it imposes on the battery Pressure Points
pack. To evenly distribute the pressure on the skin,
the rim must be a fluid-filled bladder. A semi-
fluid silicone gel, however, has mechanical
characteristics closer to those of live tissues and
interfaces better with the torso. We found that
the conforming cushion effect of a gel-filled
rim complies best with the motion of the torso
during routine activities, while evenly distribut-
ing the pressure and accommodating individ-
ual variations in surface contour. This feature
of the system allows it to prevent both the
development of localized pressure points and
breaks in skin contact that can lead to loss of
vacuum.
Maintenance of Low-Level Vacuum Seal
To prevent air leaks and to maintain the low
vacuum seal with a low contact pressure, the
contact surface of the rim against the skin must
FIG. 2. The contact area of the rim with the skin sur-
be sticky. Loss of the stickiness led to repeated
rounding each breast must be relatively large to distribute the loss of vacuum, excessive activity of the pump,
pressure caused by the distraction counter-force. F(out) in- and a rapid power drain of the battery pack. A
dicates the resultant force of distraction applied to the breast layer of tacky, hypoallergenic silicone gel was
enclosed under the dome aperture area (A) by the vacuum added to achieve the proper seal effect.
pressure [P(v)]; F(in), the resultant counter-force inflicting
a pressure [P(s)] to the chest wall under the rim area (R). To Methods
balance the forces and to optimize the pressure to the highest
continuously tolerable level, the rim contact area (R) is ap- After Institutional Review Board (IRB) ap-
proximately equal to the dome aperture area (A). proval for the study and after obtaining written
Vol. 105, No. 7 / NONSURGICAL BREAST ENLARGEMENT 2503

FIG. 3. Distraction of the breast inside the dome stretches the skin. If the peripheral breast
skin is held firmly at the inner edge of the rim, a strong shear force [F(s)] combines with the
counter-force [F(in)] to cause skin damage (left). A rim that can deflect inward allows peripheral
chest wall skin to move inside the dome. This recruitment of skin reduces the amount of breast
skin stretch and the potentially damaging shear force [F(s)] (right).

consent, we enrolled 17 female volunteers at technique (modified from Campaigne et al.25 )

one study center. Inclusion criteria were an age and the Grossman-Roudner device,26 and the
between 18 and 40 years; general good health; difference in chest circumference measured at
high motivation for breast enlargement; cur- the level of the nipple versus the level of the
rent cup size of AA, A, or small B; agreement to inframammary fold. Body weight was closely
use medically accepted birth control for the monitored, and a larger than 5 percent varia-
duration of the study; and willingness to com- tion was grounds for study withdrawal. Plaster
ply with stringent protocol requirements. Vol- moulages of the torso were made at baseline, at
unteers were excluded from the study on the the end of the treatment period, and at the
basis of a positive urine pregnancy test; ongo- 4-week follow-up visit.
ing lactation; history of breast surgery, disease, Baseline magnetic resonance imaging (MRI)
cyclic engorgement, trauma, or pain; presence examinations with breast surface coils were
of a breast mass; severe ptosis of the breast; conducted before treatment began between
history of chronic dermatitis; and any hor- days 6 and 14 of the menstrual cycle. Axial and
monal therapy besides birth control pills. coronal views were obtained using T1 and
Participants were required to wear the STIR imaging sequences.27–29 At least 1 week
breast-enlargement system for 10 to 12 hours after completion of the treatment phase (and
per day every day for 10 weeks. Failure to use in the same phase of the menstrual cycle as the
the system for 2 consecutive days or for more first MRI), a follow-up breast MRI was obtained
than 3 separate days in any given 2-week period using the same imaging planes and pulse se-
was compensated for by an extra 2 weeks of quences. The paired scans were read and
applied use. Progress was monitored by visits at graded by a radiologist experienced in MRI
1, 3, 5, 7, and 10 weeks into treatment, and at who was blinded as to the sequence of the
1, 4, 8, 20, and 30 weeks after the termination examinations.
of treatment. Breast dome size of the system Three of the women who experienced the
was increased as needed to accommodate the greatest amount of growth were recalled 26
growing breasts. Breast size was monitored dur- weeks after treatment for random needle biop-
ing visits by standardized photographs, volume sies of the breast. The tissue was processed for
measurement using both a bead displacement histologic examination with hematoxylin and
2504 PLASTIC AND RECONSTRUCTIVE SURGERY, June 2000
eosin staining; the samples were read by a both measures). This is a net increase of 55
breast pathologist. percent over the initial breast volume (Fig. 4).
All participants also completed self-esteem Every participant experienced a net volume
and opinion-of-use surveys both before and af- increase (range, 15 to 115 percent of initial
ter treatment. To obtain direct feedback on volume). The volume changes were paralleled
the change produced by the use of the system, by a 5.3 ⫾ 0.7 cm increase in the difference
a simple linear numeric scale was designed. In between circumferential chest measurements
a 10-item satisfaction questionnaire, the partic- at the nipple level versus the inframammary
ipants gave responses using a five-point scale, fold (p ⬍ 0.01 compared with the baseline
ranging from “strongly disagree” to “strongly difference; Fig. 5). Volume measurement of
agree” with each statement. the plaster moulages showed the same increase
Changes in breast volume, chest measure- in breast size as measured by the bead displace-
ment, and body weight were statistically com- ment and the Grossman-Roudner device. At
pared with a paired t test. Changes in responses last follow-up, the participants were still wear-
to the questionnaire were analyzed with a non- ing new bras at least one cup size larger than
parametric (Wilcoxon sign rank) test. their pretreatment size.
The breast volume enlargement was not as-
RESULTS
sociated with an increase in body weight dur-
Seventeen women were enrolled, and 12 ing the course of study. Quite the contrary, a
completed the study; the five withdrawals were statistically significant trend toward body
due to protocol noncompliance. All partici- weight reduction occurred in the months after
pants found the breast-enlargement system
comfortable to wear although somewhat obtru-
sive under tight-fitting clothes. Most preferred
to wear it at home and slept with it at night.
They used the system an average of 10.4 hours
per day. It took an average of 14.7 weeks
(range, 10 to 18 weeks) to complete the effec-
tive 10-week treatment period because of set-
backs from the occasional inability to use the
device. At some point during the course of
expansion, some women experienced transient
and completely reversible dysesthesia of the
nipple that was probably caused by stretch of
the sensory nerves. There were no serious ad-
verse events, pressure sores, or significant der-
matological reactions attributable to the use of
the device.
The average breast volume (mean ⫾ stan-
dard error of the mean) of the participants
before initiating treatment was 192 ⫾ 64 ml or
203 ⫾ 73 ml, as measured by bead displace-
ment or the Grossman-Roudner device, respec-
tively. The volume immediately after the 10-
week treatment was 347 ⫾ 67 ml or 344 ⫾ 62
ml (bead displacement or Grossman-Roudner
device methods, respectively; p ⬍ 0.0001 com-
pared with baseline volumes for both mea-
sures). An immediate recoil effect (loss of en-
largement) occurred over the first follow-up
week; this was followed by a plateau and a net
stable volume increase of 103 ⫾ 35 ml or 111 ⫾ FIG. 4. Breast volume increase over the course of the 10
weeks of treatment and the additional 30 weeks of follow-up.
46 ml (bead displacement or Grossman- (Above) Measurements obtained by the bead displacement
Roudner device methods, respectively; p ⬍ method. (Below) Measurements obtained by the Grossman-
0.0001 compared with baseline volumes for Roudner device. Data are means ⫾ standard deviations.
Vol. 105, No. 7 / NONSURGICAL BREAST ENLARGEMENT 2505
breasts were imaged within hours of device
application, a higher water tissue content was
apparent (unpublished data).
Histologic examination of the biopsies taken
from six of the most enlarged breasts revealed
unremarkable, normal breast tissue paren-
chyma and architecture.
DISCUSSION
Breast size in the healthy premenopausal
adult woman is stable and varies only with preg-
nancy, hormonal intake, body weight fluctua-
tion and, to a smaller extent, during the men-
FIG. 5. Breast enlargement as determined by the increase strual cycle. Our study carefully controlled for
in the difference between the chest circumference at the all these factors. The participants were initially
nipple level and that measured at the inframammary fold screened with a pregnancy test, and they used
(mean ⫾ standard deviation).
contraceptive measures throughout. If the
women were on birth control pills before en-
treatment (p ⬍ 0.02; Fig. 6), possibly as a re- rollment, they kept taking the pills and no
sponse to improved self-image. additional hormonal treatment was given.
All women expressed satisfaction with the Body weight was carefully monitored during
outcome. This was reflected by changes in the the study and, in fact, significantly decreased
baseline scores on the questionnaire after com- while breast size increased. The most consis-
pleting the treatment (Table I). The women tent decrease in body weight was seen after the
felt their breasts were lifted after treatment and participants had completed their initial wear-
that the enlargement had a natural and aes- ing period. Except for the interim recordings,
thetically pleasing shape (Figs. 7 through 9). all breast volume measurements were made
In each case, the posttreatment MRI con- during the first phase of the menstrual cycle.
firmed an overall increase in breast size. No Therefore, although the study could not be
space-occupying lesions, breast cysts, or other blinded and it included no controls, the sub-
masses were present on pretreatment or post- stantial breast growth measured should be at-
treatment MRI scans. The T1 images showed a tributed to the use of the device.
proportionate amount of fat and fibroglandu- To our knowledge, this noninvasive external
lar tissue, with preservation of the native archi- distraction device is the only documented
tecture of the breast. STIR images of posttreat- method of nonsurgical, nonpharmacologic
ment breast tissue did not demonstrate breast enlargement. The nonsurgical system
substantial increased signal intensity, i.e., there was well tolerated, with no complications. All
was no evidence of significant breast edema patients who completed the effective 10-week
(Fig. 10). However, when the peak-enlarged treatment course achieved a significant in-
crease in breast size, even after posttreatment
recoil. This increase in breast size was still ob-
served at 30 weeks posttreatment. The growth
appears to be evenly distributed among all tis-
sue elements of the breast, with preservation of
the normal composite tissue architecture.
Stretch is good for a cell; it is the mechanism
involved in normal tissue growth, regenera-
tion, and homeostasis.15 The phenomenon of
stretch-induced tissue growth is widely preva-
lent, and it has been studied in vitro and in vivo
for many years. Cells in culture will respond to
longitudinal stretching by an increased mitotic
FIG. 6. Relative change in body weight during the study rate and by realigning their shape and cytoskel-
compared with the enrollment weight (mean ⫾ standard eton parallel to the direction of force.30 –32 Ex-
deviation). perimental studies of tissue expansion have
2506 PLASTIC AND RECONSTRUCTIVE SURGERY, June 2000
TABLE I
Change in Response to Self-Esteem Questionnaire

Change in
Questionnaire Statement Score p

1. I like my breasts just as they are now. 1.8 ⬍0.01

2. I like the size of my breasts just as they are now. 1.9 ⬍0.01
3. I am satisfied with feedback from others concerning the change in my breasts. 0.9 NS
4. I am now concerned that my body image is not attractive unless I have larger breasts. ⫺0.7 NS
5. I now have difficulty wearing light clothes or bikini beach wear due to my small breast size. ⫺1.0 NS
6. I now feel comfortable wearing beach wear. 1.4 ⬍0.05
7. I have difficulty showing my breasts during intimate interaction with my partner. ⫺0.6 NS
8. I enjoy looking at myself in the mirror in the nude. 1.1 ⬍0.05
9. I feel embarrassed when I or someone else looks at my breasts. ⫺1.2 NS
10. I feel embarrassed when someone else touches my breasts. ⫺1.3 ⬍0.05
Subject responses to the 10 questions ranged from 1 to 5: 1 indicated “least likely to be true” and 5, “most likely to be true.” The numbers in the table indicate
the difference in responses for each question (after ⫺ before treatment). Numbers greater than zero identify questions for which the participant response was more
strongly in agreement after treatment. Statistical comparisons for each question were made with the Wilcoxon sign rank test. p ⬎ 0.05 was considered not significant
(NS).

shown fibroblast changes consistent with the more recently, to restore deficient mandibles12
production of new extracellular matrix.33,34 All and entire faces13 to normal. Ilizarov9,10 dem-
types of tissues studied grow and regenerate onstrated in the laboratory and in the clinic
normal tissue when subjected to mechanical that distraction is the only known means of
stretch, including the skin,6,35,36 bones,9 –12 bow- inducing true tissue regeneration in the adult.
els,37 lungs,38 urogenital viscera,39 blood ves- How the tissue responds to external distrac-
sels,30,40 nerves,40,41 skeletal muscle,10,42 cardiac tive forces is not fully known. Cells experience
tissue,43 and smooth muscles.30 A number of a variety of forces throughout their lifetime.
widely used medical devices rely on this prin- They sense the balance of mechanical forces
ciple to generate skin for wound closure,7,44,45 that surround them and translate changes into
to reconstruct the breast after a mastecto- biochemical signals by a mechanism called
my,46,47 to elongate entire extremities9 and, mechanotransduction.15 Cells are mechanically

FIG. 7. Representative photographs of a woman in the study. Left, frontal views; right, oblique
views; above, views before treatment; below, views after treatment with the nonsurgical breast-
enlargement system at 30 weeks of follow-up.
Vol. 105, No. 7 / NONSURGICAL BREAST ENLARGEMENT 2507

FIG. 8. Representative photographs of a woman in the study. Left, frontal views; right, oblique
views; above, views before treatment; below, views after treatment with the nonsurgical breast-
enlargement system at 30 weeks of follow-up.

linked to other cells and to the extracellular til the gap is filled and the normal balance is
matrix through their cytoskeleton and its sur- restored again.15
face receptor system.15 Integrins are thought to In view of the substantial societal demand for
be the transmembrane receptor link between breast enlargement (16 to 19 million women in
the mechanical deformation of the extracellu- the United States), the application of this phe-
lar matrix caused by external forces and the nomenon of stretch-induced tissue growth to
resultant internal cytoskeletal conformational the breast was bound to be forthcoming. By
response. Mechanical stretching of the extra- taking into account biomechanical and physi-
cellular matrix induces integrin clustering and ologic constraints and after extensive prototyp-
ligand binding to form macromolecular scaf- ing, we discovered four critical design features
folds called focal adhesion complexes, which and determined the most effective and safe
mechanically link the extracellular matrix with distraction pressure. The nonsurgical system
the cytoskeleton and bring the tensional forces used in this study applies an external, low-level,
into balance.48 This balancing of the mechan- sustained traction that is effective and well tol-
ical forces involves tensegrity, an architectural erated. To stimulate tissue growth, however,
system in which structures stabilize themselves the distraction must be continuous and sus-
by counteracting the forces of compression tained over a prolonged period of time. This is
and tension.49 The formation of focal adhesion the reason why failure to continuously use the
complexes also mediates the stimulus-coupling device leads to setbacks and necessitates addi-
response with the activation of kinases, ion tional compensatory wear time. Although the
channels, and growth factor receptors.48,50 –53 numbers are too small to evaluate statistically,
Mechanically induced rearrangements of the it seems that the greatest effect was observed in
cellular cytoskeleton are also directly linked to the participants who used the system the most
the nucleus to initiate cell division. This pro- intensively (hours/day) and the most continu-
cess is the subject of intense research and many ously (missed no days).
authoritative reviews.14 –17,48 –54 From a teleolog- The tissue growth achieved in this study is
ical viewpoint, whenever stretched and de- generated by the same basic mechanism of
formed, cells in the tissue sense the need to physical distraction as occurs with surgically
spread; they then respond by proliferating un- implanted bone lengthening and tissue expan-
2508 PLASTIC AND RECONSTRUCTIVE SURGERY, June 2000

FIG. 9. Representative photographs of a woman in the study. Left, frontal views; right, oblique
views; above, views before treatment; below, views after treatment with the nonsurgical breast-
reduction system at 30 weeks of follow-up.

sion devices.6 –13 Subcutaneously implanted tis- count for some noticeable growth after a short
sue expanders, silicone bladders that are filled period of use. This early volume increase in-
with saline, act on the overlying skin and asso- cludes no true tissue growth and is totally re-
ciated tissues. There is an initial elastic versible. It is only after continued use and sus-
(stretch) response, followed by the induction tained stretch that true tissue growth is
of mitosis35 and a remodeling of the connective stimulated. The stretched cells respond to the
tissue extracellular matrix36 in the stretched sustained deformation by undergoing mitosis
tissue. As the tissue expands (with a resultant and the deposition of additional extracellular
reduction in tension), more saline is added to matrix. With sustained use over a number of
the bladder to continually exert tension and weeks, the incremental component of true tis-
cause growth. The Ilizarov bone-lengthening sue growth adds up on top of the reversible
system and related devices work similarly: a elastic deformation and extracellular fluid ac-
transverse osteotomy sets up bone callus forma- cumulation. This accounts for the marked
tion, which is then gradually distracted, allow- peak enlargement seen at the end of the treat-
ing the newly forming bone to be extracted ment phase and the recoil observed 1 week
(grown) at the callus site.9 –11 Associated soft later as both old and new tissues return to their
tissues (muscle, nerve, vessels, etc.) are also resting state. At the end of a 10-week cycle of
distracted, with an initial stretch and deforma- use, breast volume is expected to approxi-
tion response followed by true tissue growth, mately double, with half of this gain remaining
which occurs evenly along the distracted area, as long-term growth. The final tissue seems to
not just at the level of the callus.40 be stable over a 30-week follow-up, and it has a
Under the effect of the external three- normal histologic appearance.
dimensional pull applied by the nonsurgical The MRI evaluation was used to gauge
breast-enlargement system, the breast tissue whether breast augmentation changed the ra-
goes through several stages in its expansion. tio and distribution of fatty and fibroglandular
During the early phase, fibroelastic fibers, ini- tissue (as depicted on T1 images) and to de-
tially loosely spaced around fat and fibroglan- termine whether there was increased water
dular cells, are stretched. Elastic deformations, content or inflammation after treatment (seen
along with some increased water edema ac- on STIR images). In all cases, the increase in
Vol. 105, No. 7 / NONSURGICAL BREAST ENLARGEMENT 2509

FIG. 10. T1-weighted coronal MRIs of a participant at baseline (above) and after treatment
(below) show an increase in size with a proportionate increase in fatty tissue (white) and fi-
broglandular tissue (gray). Note that there is no change in the distribution or appearance of the
fatty and fibroglandular components.

the size of breasts after treatment correlated rectly transmitted through the loose skin to
with nearly equal increases in fatty and fi- cause an enlargement of the tighter contents
broglandular tissue, without discernible before skin enlargement occurs. It is expected,
changes in the tissue architecture. Increased however, that gravity acting on the now-larger
signal intensity on STIR images was not appar- breast will naturally tend to cause some ptosis
ent; this suggests that increased water content with time.
(due to edema and/or inflammation) was not Another major concern about this breast-
present. Although contrast-enhanced MRI is enlargement system is whether its application
more sensitive in the detection of breast cancer stimulates or accelerates latent breast cancer.
and benign processes, such as dysplasia and Although the number of participants and the
inflammatory breast disease,28,29 the noncon- time course for follow-up are too small to de-
trast images did not demonstrate any overt termine this, several related findings do not
changes that would suggest the development of support a cancer-inducing/stimulating effect
disease when comparing posttreatment and by tension-induced tissue growth. Mechanical
pretreatment images. forces are not known to be carcinogens. The
Concern that stretching the breast would force used by the device is trivial compared
accentuate any degree of ptosis was not borne with the ones that constantly act upon the
out by the study results. The most noticeable body. This vacuum pressure of 20 mmHg rep-
initial impression by all the participants was resents a 2.5 percent drop in atmospheric pres-
that of a breast fill and lift. In the ptotic breast, sure. It is equivalent to the pressure change
there is a discrepancy between the loose skin experienced before a storm or when climbing
envelope and the relatively smaller contents. to the top of a high tower, and it is 4 to 7 times
The forces of distraction, therefore, are di- less than the pressure change experienced in-
2510 PLASTIC AND RECONSTRUCTIVE SURGERY, June 2000
side the cabin of a commercial aircraft. The monally mediated, whereas the growth of the
total mechanical pull exerted by the device on overlying skin is induced by secondary mechan-
the breast is approximately equivalent to the ical stretch. After weight loss or deliveries, the
force exerted by gravity on a large, 2-kg breast. mechanical stretch and the supporting scaffold
This amount of force leads to downward (uni- regress, while skin growth often remains as a
dimensional) growth of the unsupported cosmetic problem.
larger breast, an effect well-known to plastic Plastic surgeons are also familiar with the
surgeons performing reduction mammaplas- ptosis that follows the removal of a breast im-
ties. Numerous epidemiologic studies have plant. Here, sustained stretch by the implant
failed to reveal any increased cancer incidence induces the growth of the tissue in the breast
in the larger, heavier breasts that are subjected, envelope. The resultant ptosis will not recoil
over a lifetime, to a mechanical stretch similar with time, and the enlarged tissues will not
to that of the nonsurgical breast-enlargement shrink back. Similarly, when a tissue expander
system.55–58 This experiment of nature proves is rapidly inflated and removed, no significant
that mechanical forces acting on the breast are tissue growth occurs, only reversible stretch
not carcinogenic. and recruitment. However, when an inflated
Ilizarov devices do not have an associated expander is kept for a few months, its removal
cancer induction with their use in distraction is followed by permanent wrinkling of the skin.
of the extremities.9 Skin is the most cancer- That soft tissues recoil after losing their struc-
prone organ in the body, yet skin expanders tural support is not universally true. It is the
have been used for decades without any report presence of inflammation that may cause un-
of cancer arising in the expanded skin.7,8,59,60 derlying scarring and some degree of contrac-
Furthermore, these tissue expanders have
tion. The nonsurgical breast-enlargement sys-
been applied in breast reconstruction after
tem was designed to specifically avoid any
mastectomy, stretching the residual breast tis-
significant tissue inflammation.
sue that has a high likelihood of tumor recur-
In summary, this system offers women a
rence (it is well accepted that even the most
radical of the mastectomies leaves some breast means of enlarging their breasts without the
tissue behind). Yet more than 20 years of ex- pain and risk of surgery or the perceived long-
perience in thousands of women has con- term health risks associated with surgical im-
firmed that breast reconstruction with tissue plants. The process is slow and gradual; the
expansion does not increase the incidence of arbitrarily chosen 10-week course does not
cancer recurrence.42,43,60 – 63 Furthermore, in an match the immediate size gain that follows the
experimental model of rat mammary carci- insertion of an average sized breast implant.
noma, tissue expansion induced engrafted tu- Additional use beyond 10 weeks is required to
mor regression and even a reduction in the achieve further growth. This gradual process
spread of visceral metastasis.64 gives women more control over the change in
We can only speculate as to why the breasts their appearance. Because the tissue growth is
seem to retain their newly gained growth after local and autogenous, the result is more natu-
the rapid elastic recoil and the loss of tissue ral looking, as opposed to the artificial appear-
edema. Aside from mechanical stretch, growth ance that can often accompany breast im-
factors and hormones can also stimulate tissue plants.
growth. The administration of these factors Roger K. Khouri, M.D.
stimulates new tissue growth; the survival of Dermatology and Plastic Surgery
this growth is critically dependent on the con- 328 Crandon Blvd., Suite 227
tinued presence of the hormonal stimulus.65– 68 Key Biscayne, Fla. 33149
The new tissues regress by apoptosis on with- [email protected]
drawal of the growth factor. Tissue growth in-
duced by mechanical stretching, however, may ACKNOWLEDGMENTS
remain, even after the withdrawal of the me-
chanical stimulus. Experiments of nature fa- The study was supported by Biomecanica, Inc., Miami,
Florida (owner of the patented Distraction Augmentation
miliar to plastic surgeons demonstrate how Mammoplasty technology). The authors thank Marita Eisen-
these two types of induced tissue growth may man-Klein, M.D., and Jurgen Holle, M.D., for assisting with
differ. With weight gain or pregnancies, growth the study, Ann Pando, Ph.D., for designing the psychological
of the adipose tissue or of the uterus is hor- questionnaire, and Brian Cooley for editing the manuscript.
Vol. 105, No. 7 / NONSURGICAL BREAST ENLARGEMENT 2511
REFERENCES surement and the effect of externally applied pres-
1. Cosmetic Surgery Telephone Survey, National Clearing- sure. Arch. Phys. Med. Rehabil. 62: 390, 1981.
house of Plastic Surgery Statistics, 1998. 24. Zhang, M., and Roberts, V. C. The effect of shear forces
externally applied to the skin surface on underlying
2. Hase/Shannen Quantitative Market Research, 1999.
tissues. J. Biomed. Eng. 15: 451, 1993.
3. American Society of Plastic and Reconstructive Surgeons.
25. Campaigne, B. N., Katch, V. L., Freedson, P., and Sady,
Plastic Surgery News, June 1999.
S. Measurement of breast volume in females: De-
4. Burnett, W. Yank meets native. Natl. Geogr. 88: 105,
scription of a reliable method. Ann. Hum. Biol. 6: 363,
1945.
1979.
5. Weeks, G. S. Into the heart of Africa. Natl. Geogr. 110:
26. Grossman, A. J., and Roudner, L. A. A simple means for
257, 1956.
accurate breast volume determination. Plast. Reconstr.
6. Neumann, C. G. The expansion of an area of skin by
Surg. 66: 851, 1980.
progressive distention of a subcutaneous balloon.
27. Harms, S. E., and Flamig, D. P. MR imaging of the
Plast. Reconstr. Surg. 19: 124, 1957.
breast. J. Magn. Reson. Imaging 3: 277, 1993.
7. Manders, E. K., Schenden, M. J., Furrey, J. A., Hetzler,
28. Gribbestad, I. S., Nilsen, G., Fjosne, H. E., Kvinnsland, S.,
P. T., Davis, T. S., and Graham, W. P. Soft-tissue
Haugen, O. A., and Rinck, P. A. Comparative signal
expansion: Concepts and complications. Plast. Recon-
intensity measurements in dynamic gadolinium-en-
str. Surg. 74: 493, 1984.
hanced MR mammography. J. Magn. Reson. Imaging 4:
8. Marcus, J., Horan, D. B., and Robinson, J. K. Tissue
477, 1994.
expansion: Past, present, and future. J. Am. Acad. Der-
29. Orel, S. G., Schnall, M. D., LiVolsi, V. A., and Troupin,
matol. 23: 813, 1990.
R. H. Suspicious breast lesions: MR imaging with
9. Ilizarov, G. A. Clinical application of a tension-stress radiologic-pathologic correlation. Radiology 190: 485,
effect for limb lengthening. Clin. Orthop. 250: 8, 1990. 1994.
10. Ilizarov, G. A. The tension-stress effect on the genesis 30. Wilson, E., Mai, Q., Sudhir, K., Weiss, R. H., and Ives,
and growth of tissue. Clin. Orthop. 238: 249, 1989. H. E. Mechanical strain induces growth of vascular
11. Yasui, N., Kojimoto, H., Shimizu, H., and Shimomura, Y. smooth muscle cells via autocrine action of PDGF.
The effect of distraction upon bone, muscle, and peri- J. Cell Biol. 123: 741, 1993.
osteum. Orthop. Clin. North Am. 22: 563, 1991. 31. Iba, T., Maitz, S., Furbert, T., et al. Effect of cyclic
12. McCarthy, J. G., Schreiber, J., Karp, N., Thorne, C. H., stretch on endothelial cells from different vascular
and Grayson, B. H. Lengthening the human man- beds. Circ. Shock 35: 193–198, 1991.
dible by gradual distraction. Plast. Reconstr. Surg. 89: 1, 32. Brunette, D. M. Mechanical stretching increases the
1992. number of epithelial cells synthesizing DNA in cul-
13. Polley, J. W., and Figueroa, R. A. Management of severe ture. J. Cell Sci. 69: 35, 1984.
maxillary deficiency in childhood and adolescence 33. Pasyk, K. A., Austad, E. D., McClatchey, K. D., and Cherry,
through distraction osteogenesis with an external, ad- G. W. Electron microscopic evaluation of guinea pig
justable, rigid distraction device. J. Craniofac. Surg. 8: skin and soft tissues “expanded” with a self-inflating
181, 1997. silicone implant. Plast. Reconstr. Surg. 70: 37, 1982.
14. Chen, C. S., Mrksich, M., Huang, S., Whitesides, G. M., 34. Mustoe, T. A., Bartell, T. H., and Garner, W. L. Physical,
and Ingber, D. E. Geometric control of cell life and biomechanical, histologic, and biochemical effects of
death. Science 276: 1425, 1997. rapid versus conventional tissue expansion. Plast. Re-
15. Ingber, D. E. Tensegrity: The architectural basis of cel- constr. Surg. 83: 687, 1989.
lular mechanotransduction. Annu. Rev. Physiol. 59: 35. Olenius, M., Dalsgaard, C.-J., and Wickman, M. Mitotic
575, 1997. activity in expanded human skin. Plast. Reconstr. Surg.
16. Sachs, F. Biophysics of mechanoreception. Membr. Bio- 91: 213, 1993.
chem. 6: 173, 1986. 36. Beauchene, J. G., Chambers, M. M., Peterson, A. E., and
17. Ruoslahti, E. Stretching is good for a cell. Science 276: Scott, P. G. Biochemical, biomechanical, and phys-
1345, 1997. ical changes in the skin in an experimental animal
18. The Princess Breast Developer. In Sears, Roebuck, and Co. model of therapeutic tissue expansion. J. Surg. Res. 47:
Catalog. Chicago, Ill.: Sears, Roebuck, and Co., 1902. 507, 1989.
19. Hayes, H., Jr. Quackery and Nostrums. In An Anthology 37. Kimura, K., and Soper, R. T. A new bowel elongation
of Plastic Surgery. Rockville, Md.: Aspen Publishers, technique for the short bowel syndrome using the
1986. Pp. 163–175. isolated bowel segment Iowa models. J. Pediatr. Surg.
20. Xakellis, G. C., Frantz, R. A., Arteaga, M., and Meletiou, 28: 792, 1993.
S. T. A comparison of changes in the transcutaneous 38. Rannels, D. E. Role of physical forces in compensatory
oxygen tension and capillary blood flow in the skin growth of the lung. Am. J. Physiol. 257: L179, 1989.
with increasing compressive weights. Am. J. Phys. Med. 39. Ikeguchi, E. F., Stifelman, M. D., and Hensle, T. W. Ure-
Rehabil. 70: 172, 1991. teral tissue expansion for bladder augmentation.
21. Holloway, G. A., Daly, C. H., Kennedy, D., and Chimos- J. Urol. 159: 1665, 1998.
key, J. Effects of external pressure loading on human 40. Ippolito, E., Peretti, G., Bellocci, M., et al. Histology and
skin blood flow measured by 133Xe clearance. J. Appl. ultrastructure of arteries, veins, and peripheral nerves
Physiol. 40: 272, 1989. during limb lengthening. Clin. Orthop. 308: 54, 1994.
22. Frantz, R. A., and Xakellis, G. C. Characteristics of skin 41. Skoulis, T. G., Lovice, D., von Fricken, K., and Terzis, J. K.
blood flow over the trochanter under constant, pro- Nerve expansion: The optimal answer for the short
longed pressure. Am. J. Phys. Med. Rehabil. 68: 272, nerve gap. Behavioral analysis. Clin. Orthop. 314: 84,
1989. 1995.
23. Newson, T. P., and Pearcy, M. J. Skin surface PO2 mea- 42. Kim, K.-H., Hong, C., and Futrell, J. W. Histomorpho-
2512 PLASTIC AND RECONSTRUCTIVE SURGERY, June 2000
logic changes in expanded skeletal muscle in rats. volumes in cancer of the breast. Br. J. Cancer 29: 270,
Plast. Reconstr. Surg. 92: 710, 1993. 1974.
43. Sadoshima, J., and Izumo, S. The cellular and molec- 57. Tavani A, Pregnolato A, La Vecchia C, Negri E, Favero A,
ular response of cardiac myocytes to mechanical Franceschi S. Breast size and breast cancer risk. Eur.
stress. Annu. Rev. Physiol. 59: 551, 1997. J. Cancer Prev. 5: 337, 1996.
44. Mingli, Z., Dawei, W., and Lan, H. The application of 58. Hsieh, C.-C., and Trichopoulos, D. Breast size, hand-
skin external expander to postburn advanced scar edness, and breast cancer risk. Eur. J. Cancer 27: 131,
contracture. Plast. Reconstr. Surg. 96: 1600, 1995. 1991.
45. Blomqvist, G., and Steenfos, H. A new partly external 59. Georgiade, G. S., Riefkohl, R., Cox, E., et al. Long-term
device for extension of skin before excision of skin clinical outcome of immediate reconstruction after
defects. Scand. J. Plast. Reconstr. Surg. Hand Surg. 27: mastectomy. Plast. Reconstr. Surg. 76: 415, 1985.
179, 1993. 60. Antonyshyn, O., Gruss, J. S., Mackinnon, S. E., and Zuker,
46. Slavin, S. A., and Colen, S. R. Sixty consecutive breast R. Complications of soft tissue expansion. Br. J. Plast.
reconstructions with the inflatable expander: A criti- Surg. 41: 239, 1988.
cal appraisal. Plast. Reconstr. Surg. 86: 910, 1990. 61. Johnson, C. H., van Heerden, J. A., Donohue, J. H., et al.
47. Chisholm, E. M., Marr, S., Macfie, J., Broughton, A. C., Oncologic aspects of immediate breast reconstruction
and Brennan, T. G. Post-mastectomy breast recon- following mastectomy. Arch. Surg. 124: 818, 1989.
struction using the inflatable tissue expander. Br. J. 62. Badellino, F., Canavese, G., Catturich, A., et al. The
Surg. 73: 817, 1986. impact of reconstructive surgery in breast cancer. Ann.
48. Ingber, D. Integrins as mechanochemical transducers. N. Y. Acad. Sci. 698: 219, 1993.
Curr. Opin. Cell Biol. 3: 841, 1991. 63. O’Brien, W., Hasselgren, P.-O., Hummel, R. P., et al.
49. Ingber, D. E. Life’s architecture: Cells grow with Comparison of postoperative wound complications
“tensegrity.” Sci. Am. 1: 2, 1998. and early cancer recurrence between patients under-
50. Sachs. F. Mechanical transduction by membrane ion going mastectomy with or without immediate breast
channels: a mini review. Mol. Cell Biochem. 104: 57, reconstruction. Am. J. Surg. 166: 1, 1993.
1991. 64. Ramasastry, S. S., Weinstein, L. W., Zerbe, A., Narayanan,
51. Watson, P. A. Function follows form: Generation of in- K., LaPietra, D., and Futrell, J. W. Regression of local
tracellular signals by cell deformation. FASEB J. 5: and distant tumor growth by tissue expansion: An
2013, 1991. experimental study of mammary carcinoma 13,762 in
52. Vandenburgh, H. H. Mechanical forces and their sec- rats. Plast. Reconstr. Surg. 87: 1, 1991.
ond messengers in stimulating cell growth in vitro. 65. Khouri, R. K., Hong, S. P., Deune, E. G., et al. De novo
Am. J. Physiol. 262: R350, 1992. generation of permanent neovascularized soft tissue
53. Chicurel, M. E., Chen, C. S., and Ingber, D. E. Cellular appendages by platelet-derived growth factor. J. Clin.
control lies in the balance of forces. Curr. Opin. Cell Invest. 94: 1757, 1994.
Biol. 10: 232, 1998. 66. Deune, E. G., Hong, S. P., Tarpley, J. E., et al. Long-term
54. Thoumine, O., Nerem, R. M., and Girard, P. R. Oscil- stability and usefulness of tissue generated by growth
latory shear stress and hydrostatic pressure modulate factors. Surg. Forum 44: 693, 1993.
cell-matrix attachment proteins in cultured endothe- 67. Khouri, R. K., Koudsi, B., Deune, E. G., et al. Tissue
lial cells. In Vitro Cell Dev. Biol. Anim. 31: 45, 1995. generation with growth factors. Surgery 114: 374, 1993.
55. Wynder, E. L. Identification of women at high risk for 68. Brown, D. M., Miller, J. A., Pasia, E. N., and Khouri, R. K.
breast cancer. Cancer 24: 1235, 1969. Soft tissue augmentation using lipogenic growth fac-
56. Katariya, R. N., Forrest, A. P., and Gravelle, I. H. Breast tors. Plast. Surg. Forum 18: 236, 1995.