CONTINUING

MEDICAL EDUCATION

Treatment of cellulite
Part I. Pathophysiology
Misbah H. Khan, MD,a Frank Victor, MD,a Babar Rao, MD,a and Neil S. Sadick, MDb
Somerset, New Jersey, and New York, New York
Cellulite is a topographic skin change that is nearly ubiquitous in postpubertal women. Treatment remains
elusive. The various treatments currently available are only partially or temporarily effective. Newer
therapeutic modalities continue to evolve without much understanding of the complex nature of cellulite.
The successful treatment of cellulite will ultimately depend upon our understanding of the pathophysiology of cellulite adipose tissue. Part I of this two-part series on cellulite reviews how the concept and
perception of cellulite has evolved over time and its proposed etiologies. The article also focuses on the
physiology of human adipose tissue, particularly regarding cellulite. ( J Am Acad Dermatol 2010;62:361-70.)
Learning objectives: After completing this learning activity, participants should be able to differentiate
adipocyte physiology between the cellulite-prone areas and other parts of the body, understand the
spectrum of conventional and interventional treatment modalities available, including their modes of action
and controversies that dominate and question their long-term efficacy based on published studies, and
discuss newer horizons for the treatment of cellulite.
Key words: brown adipose tissue; cellulite; subcutaneous fat; white adipose tissue.

Key points
d

Cellulite is an architectural disorder of human adipose tissue

Cellulite is characterized by the padded and
nodular appearance of skin in celluliteprone areas, such as the posterolateral thigh
in postpubertal females
Fundamental knowledge regarding its pathophysiology is lacking

From the Departments of Dermatology at Robert-Wood Johnson

University Hospital,a University of Medicine and Dentistry New
Jersey, Somerset, and Weill Medical College of Cornell University,b New York.
Funding sources: None.
Conflicts of interest: Dr Sadick and the Sadick Research Group
work with Syneron on clinical studies of their equipment
seeking approval by the US Food and Drug Administration. In
exchange, Sadick Dermatology receives discounted equipment.
The other authors, editors, and peer reviewers have no relevant
financial relationships to declare.
Reprint requests: Misbah H. Khan, MD, or Neil S. Sadick, MD,
Sadick Dermatology and Sadick Research Group, 911 Park
Ave, Ste 1A, New York, NY 10075. E-mail: khanmisbah6@gmail.
com; nssdern@sadickdermatology.com.
0190-9622/$36.00
2009 by the American Academy of Dermatology, Inc.
doi:10.1016/j.jaad.2009.10.042

Abbreviations used:
a-2 AR:
BAT:
BMI:
GAG:
LPLL:
MRI:
UCP-1:
WAT:

a-2 adrenergic receptor

brown adipose tissue
body mass index
glycosoaminoglycan
lipoprotein lipase
magnetic resonance imaging
uncoupling protein-1
white adipose tissue

ellulite is defined as a localized metabolic

disorder of subcutaneous tissue that provokes an alteration in the female body
shape. It presents as a modification of skin topography evident by skin dimpling and nodularity that
occurs mainly in women on the pelvic region, lower
limbs, and abdomen and is caused by the herniation
of subcutaneous fat within fibrous connective tissue,
leading to a padded or orange peelelike appearance
(Fig 1). Given the fact that occurrence of cellulite is
nearly universal in postpubertal females, it is thought
of as a female secondary sex characteristic. The term
cellulite was first reported in the French literature 150
years ago1 and was elegantly portrayed by the artists
of the 1600s (Fig 2). Synonyms include adiposis
edematosa, incipient cellulite or status protusus
cutis, full-blown cellulite or dermopanniculosis
361

362 Khan et al

J AM ACAD DERMATOL
MARCH 2010

deformans, nodular liposclerosis, gynoid lipodystromore stasis and causing alterations in the microcirphy, and edematofibrosclerotic panniculopathy.
culation of cellulite prone areas; and (5) pregnancy,
Cellulite is different from obesity. Obesity is charwhich is associated with an increase in certain
acterized by hypertrophy and hyperplasia of adipose
hormones, such as prolactin and insulin, and intissue that is not necessarily limited to the pelvis,
creased overall fluid volumeboth of these factors
thighs, or abdominal areas. In contrast, cellulite is
promote cellulite by lipogenesis and fluid retention.
most commonly found but not limited to the pelvis,
PATHOPHYSIOLOGY
thighs, and abdomen, and is a result of several
OF CELLULITE
ultrastructural, inflammatory,
Key points
histochemical, morphologic,
CAPSULE SUMMARY
d Etiologic factorsstrucand biochemical changes
tural and architectural
that produce the padded
Cellulite is a complex architectural
alterations; metabolic
and orange peel appearance
disorder with multifactorial etiologies
and biochemical differof the skin.2-5 There is no
that is prevalent in 98% of the
ences between cellulite
mortality or morbidity assopostpubertal females.
and normal fat; sexuciated with this condition,
Although the pathophysiology of
ally dimorphic skin arwhich makes it difficult to
cellulite is not fully elucidated,
chitecture; and adipose
define as a pathologic conphysiologic and biochemical properties
tissue differences in difdition.6 However, cellulite reof adipose tissue in cellulite-prone areas
ferent body partsplay
mains a common cause of
have been shown to be different from
a major role in cellulite
embarrassment to even the
subcutaneous adipose tissue elsewhere.
pathophysiology
most fit of women.
Cellulite-prone areas have a higher
d Minor factors, such as
According to Scherwitz
concentration of large metabolically
localized tissue vascularand Braun-Falco,1 it was
stable adipocytes.
ity and postinflammaAlquier and Paviot who, in
tory changes, hormonal
1920, first described cellulite
and genetic influences, lifestyle, weight gain
as a noninflammatory, complex cellular dystrophy of
or loss, etc, may also be important
the mesenchymal tissue, where defects in water
metabolism led to saturation of adjacent tissues by
Although the differences between cellulite pathinterstitial liquids as a result of traumatic, topical
ophysiology and normal fat are largely unknown,
infectious, or glandular stimuli.
several observations have been proposed. These can
Unfortunately, there has been little advancement
be subdivided into major factors, including structural
in our understanding of the pathophysiology of
and architectural alterations, metabolic and biocellulite. Research is scant, and dissention and conchemical differences between cellulite and normal
troversy are common. In the last 30 years, there have
fat (which includes abnormal hyperpolymerization
been only a handful of peer-reviewed articles in the
of connective tissue and sexually dimorphic skin
medical literature that have addressed cellulite.
architecture), adipose tissue differences in different
There is no definitive explanation for its presentation
parts of the body, and altered fatty tissue. Minor
and prevalence among 85-98% of post-pubertal
factors, such as tissue vascularity and postinflammafemales who display some degree of cellulite. This
tory changes, hormonal and genetic influences,
greatly complicates the ability to treat or improve it.
lifestyle, weight gain or loss, dietary preferences,
and pregnancy, among others, may also play a role.
d

MULTIFACTORIAL ETIOLOGIES OF
CELLULITE

Genetic predisposition plays a role in the development of cellulite. The following factors are also
important: (1) gender differencescellulite in its
classic patterns affects women exclusively; (2)
ethnicitywhite women tend to get cellulite more
than Asian women; (3) lifestylean excessive, highcarbohydrate diet provokes hyperinsulinemia and
promotes lipogenesis, leading to an increase in total
body fat content, thereby enhancing cellulite; (4)
sedentary lifestyleprolonged periods of sitting or
standing may impede normal blood flow, leading to

STRUCTURAL AND ARCHITECTURAL

DIFFERENCES BETWEEN CELLULITE AND
NORMAL FAT
Key points
d

Skin dimpling in cellulite and dermal stretch

marks, or so-called striae, may be similar
conditions with forces of distension acting
in a perpendicular versus parallel fashion
Women with cellulite have a higher percentage
of thinner, perpendicularly oriented hypodermal septae than unaffected women and men

Khan et al 363

J AM ACAD DERMATOL
VOLUME 62, NUMBER 3

Fig 1. Magnetic resonance images of adipose tissue. A,

Hypodermis of the entire thigh that appears hyperintense.
Dermis is not visible at this resolution. B, High spatial
resolution, two-dimensional image, 3 mm thick, of hypodermis on dorsal side of the thigh of a woman with
cellulite. Camper fascia separates the adipose tissue into
two layers. Fibrous tissue septae appear as thick, hypointense structures. (Reprinted with permission.11)

Connective tissue alteration

Cellulite is characterized by the presence of fatty
protrusions through the dermohypodermal junction.7
Cellulite may be caused by gender-related differences
in connective tissue that are clinically more evident
with weight gain. A study by Rosenbaum et al8
reported similar findings using ultrasonography and
full-thickness wedge biopsy of the thighs under local
anesthesia. Gross ex vivo and in vivo examination of
the thighs showed a diffuse pattern of extrusion of
underlying adipose tissue into the reticular dermis in
affected (but not unaffected) individuals, directly
correlating with clinical findings, such as dimpling
and orange peel texture of thigh skin.
However, Pierard et al9 found no correlation in
their study between the extent of these protrusions
and clinical evidence of cellulite. In a study using
autopsy specimens of 24 previously healthy women
between 28 and 39 years of age with cellulite and 11
men and four women without cellulite, the authors
revealed important distinguishing characteristics
within the microarchitecture of subcutaneous connective tissue below the dermosubcutaneous interface. They showed the presence of papillae adiposae
rising into the pits and dells on the undersurface of
the dermis with sweat glands encased within. They
found no correlation between the extent of this
finding and the clinical type and severity of cellulite
in women. In addition, the most distinguishing
feature between cellulite-prone skin and unaffected
skin was found to be uneven thickness of connective
tissue septae, showing a few a-actinepositive myofibroblasts in thicker strands. This finding correlated
with the mattress phenomenon as seen on pinching of the skin corresponding with the areas where
the septae are thicker and contain myofibroblasts. At
the site of skin dimpling at rest, they reported lumpy

and loose swellings interposed between thinner

portions of the septae. Collagen fibers formed a
delicate meshwork that was reminiscent of striae
distensea. They were found to be distributed unevenly, being either coarse and clumped or rare and
thin. Acid proteoglycans and a2-macroglobulin were
present in abundance at these sites. The study
concluded that skin dimpling in cellulite and dermal
stretch marks are similar conditions with forces of
distension acting perpendicular versus parallel, respectively.10 As such, each can coexist in the same
individual, which is a commonly noted finding. They
also found numerous blood vessels at the dermal
level compared to those poorly developed at the
subcutis without any changes in the lymphatics
(control vs affected) in the cellulite-affected areas.
Sexual dimorphic skin architecture
The anatomic hypothesis of cellulite is based on
gender-related differences in the structural characteristics of dermal and hypodermal architecture,
originally detailed by Nurnberger and Muller.7
They described herniations of fat into the dermis,
which are characteristic of female anatomy and
which were later confirmed by ultrasound imaging
as being low-density regions among denser dermal
tissue.8 Their study revealed that gender-related
differences are diffuse and not localized to the
affected areas. They reported that dermal septae
also called papillae adiposaeof affected females
are much thinner and more radially oriented than
unaffected males, therefore facilitating the extrusion
of adipose tissue into the reticular dermis. In their
study, cellulite-affected and unaffected female subjects both showed an irregular and discontinuous
dermosubcutaneous interface that was characterized
by fat protrusion into the dermis. The dermoadipose
and connective tissue interface was smooth and
continuous in male subjects.
Querleux et al11 revealed three principal orientations of septae: perpendicular, parallel, and angulated at about 458 (Figs 3 and 4). Women with
cellulite had a higher percentage of perpendicular
septae than unaffected women (P \ .001) or men
(P \.01). For the other two directions, according to
presence of cellulite, women with cellulite had a
smaller percentage of septae parallel to skin (P \
.001) and higher percentage at 458 (P \.001).

METABOLIC AND BIOCHEMICAL

FEATURES OF CELLULITE VERSUS
NORMAL FAT
Key points
d

Catecholamine-induced lipolytic responsiveness is greatest in visceral fat than in

364 Khan et al

J AM ACAD DERMATOL
MARCH 2010

Fig 2. Grade III cellulite: skin dimpling with the patient in the supine position. (Venus and
Cupid, c. 1560, by Lambert Sustris. Oil on canvas, 132 3 184 cm. On public display at Musee du
Louvre, Paris, France; printed with permission. Photograph by the author.)

Fig 3. Visualization of three-dimensional topography of skin as seen by magnetic resonance

imaging at the interface between the dermis and subcutaneous tissue. A, Woman with cellulite.
B, Unaffected woman. C, Unaffected male. Deep adipose indentations into the dermis are a
characteristic hallmark of cellulite. (Reprinted with permission.11)

abdominal subcutaneous tissue and is least

responsive in cellulite-prone areas
White and brown adipose tissue can be better defined by features other than histology
It is not clearly known whether the cellulite
is brown or white adipose tissue

Adipose tissue in gluteofemoral versus

abdominal regions
The topographic anatomy of fatty tissue includes
two layers separated by a superficial fascia. The
more external layer or areolar layer consists of
vertically oriented globular large adipocytes. The
deeper layer, known as the lamellar layer, has

horizontally arranged smaller cells with larger and

more numerous blood vessels. Women and children
tend to have a thick areolar layer,12 which in turn is
thicker in femoral regions. Fatty tissue development
during puberty is greater in women than men. This
maybe explained by the influence of estrogen,
because 17-b-estradiol stimulates the replication
of adipocytes13,14 (Fig 5; Table I). The adipocytes
in the gluteofemoral region are larger and are
influenced by female sex hormones. They are metabolically more stable and resistant to lipolysis.15 In
addition, estrogen increases the response of the
adipocytes to antilipolytic a-2 adrenergic receptors
(a-2 ARs).

J AM ACAD DERMATOL

Khan et al 365

VOLUME 62, NUMBER 3

Fig 4. Visualization of the three-dimensional architecture of fibrous septae in subcutaneous

adipose tissue as viewed by magnetic resonance imaging. A, Woman with cellulite. B,
Unaffected woman. C, Unaffected male. (Reprinted with permission.11)

Fig 5. Overview of fatty acid uptake, lipogenesis, and lipolysis: triacylglycerol (TG) hydrolysis
leads to lipolysis, whereas fatty acid (FA) uptake and TG synthesis leads to lipogenesis. AC,
Adenylate cyclase; ACS, acyl-coenzyme A synthase; 5-AMP, 5-adenosine monophosphate;
ATP, adenosine triphosphate; aP2, adipocyte binding protein; a2-AR, a2 adenoreceptor; b-AR,
b-adenoreceptor; FATP, fatty acid transport protein; FFA, free fatty acid; GI, inhibitory G
protein; GS, stimulatory G protein; Glut4, insulin-sensitive glucose transporter; glycerol-3P,
glycerol 3-phosphate; HSL, hormone sensitive lipase; LPL, lipoprotein lipase; p, phosphorylation; PKA, protein kinase A. (Data from Marks DB, Marks AD, Smith CM. Basic medical
biochemistry: a clinical approach. Philadelphia: Lippincott, Williams and Wilkins; 1996.)

The only hormones that acutely affect lipolysis in

human adipocytes are catecholamines (epinephrine
and norepinephrine, which are lipolytic) and insulin
(antilipolytic). Functionally, there are marked regional differences in both hormonal responsiveness
and metabolic activity of human adipose tissue.
Catecholamine-stimulated lipolytic responsiveness
is greater in viscera than in abdominal subcutaneous
tissue and with less responsiveness than is seen in
gluteofemoral fat cells.16,17 The regulation of lipolysis by catecholamines involves AR stimulation of
adenylate cyclase via b-ARs (b1-, b2-, and b3-ARs)
and inhibition by a-2 ARs18 (Fig 5). Both abdominal
and gluteal adipocyte cell size correlate directly with

a-2 AR density (P \.01). The fact that the ratio of a-2

AR to b-AR is higher in the gluteal region than in
abdominal adipocytes accounts for some of the
enhanced responsiveness of abdominal fat cells
versus gluteal fat cells to mixed AR agonists, such
as epinephrine and norepinephrine. In addition,
abdominal adipocytes have a greater sensitivity to
pure b-AR agonists, such as isoproteronol.19,20 These
factors are responsible for enhanced lipolysis of
abdominal adipocytes secondary to catecholamine
stimulation as compared to gluteal adipose tissue.
Adipose tissue lipoprotein lipase (LPL) correlates
directly with the adipose cell size and its affinity for
b-AR.20,21 Catecholamine-induced lipolysis, as

366 Khan et al

J AM ACAD DERMATOL
MARCH 2010

Table I. Lipolysis and lipogenesis regulatory factors

Factors

Type of effect(s)

Functions

Obesity genes
Leptin

Endocrine/paracrine

Adipostat signal to the brain, decreases food

intake, increases energy expenditure,
increased lipolysis, decreased TG
synthesis and lipolysis

Proteins of lipid metabolism

Glut-4
LPL

Transmembrane glucose transporter

Paracrine

Positive effector of lipogenesis

Hydrolyzes lipoprotein-associated fatty
acids; positive effector of lipogenesis
Antilipolysis; vasodilator
Positive effector of lipolysis
Positive effector of lipolysis
Positive effector of lipolysis
Antilipolysis; negative effector of lipolysis

Adenosine
ALBP/aP2
Perilipin
b-AR
a2-AR

Autocrine
Intracellular FA binding protein
Intracellular lipid droplet associated protein
Transmembrane catecholamine receptor
Transmembrane catecholamine receptor

ALBP/aP2, Adipocyte lipid binding protein; a2-AR, a2-adrenoreceptor; b-AR, b-adrenoreceptor; FA, fatty acid; Glut4, insulin-dependent
glucose transporter; LPL, lipoprotein lipase; TG, triacylglycerol.
Data from Wajchenberg BL. Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome. Endocrine Rev 2000;21:697-738.

measured by localized LPL release, clearly suggests

that abdominal adipocytes have an abundance of
b-AR with greater cell size in postmenopausal
women. Stimulated LPL activity in abdominal fat
cells directly correlates with greater central obesity
seen in postmenopausal women as opposed to
gynoid feminine type obesity or cellulite, which is
more prevalent in premenopausal women. Gluteal
fat cells are larger in size and richer in a-2 AR in
premenopausal women and postmenopausal
women who are undergoing estrogen replacement
therapy.
Cellulite versus normal adipose tissue:
White, brown, or both?
Adipocytes are organized in a multidepot organ,
with only one third of adipose tissue containing
mature adipocytes. The remaining two-thirds consist
of a combination of nerves, fibroblasts, and adipocyte
precursor cells, or preadipocytes. Mature adipocytes
exist as two cytotypeswhite adipose tissue (WAT)
and brown adipose tissue (BAT)that are distinguished by their color and function. WAT is yellow or
ivory and contains predominantly white adipocytes.
BAT, which appears brown, contains multilocular
brown adipocytes. Compared with WAT, BAT contains a richer vascular tree and denser capillaries in
combination with mitochondria, which accounts for
its brown color. They are both innervated by the
noradrenergic sympathetic nervous system.
BAT and WAT are histologically distinct yet interchangeable.22,23 Lipids in WAT are organized within
one large, unilocular droplet, the size of which
exceeds 50 m. White adipocytes are spherical,

allowing for maximum volume expansion within

minimal space. The nucleus is compressed to one
side because of the high lipid content. Lipids within
brown adipocytes are organized into multiple
smaller, multilocular droplets. They have higher
mitochondrial content packed with cristae within
the cytoplasm. Cells are polygonal, have centrally
placed nuclei, and are relatively smaller than WAT,
ranging from 20 to 40 m. WAT is distributed in
several anatomically distinct and separate collections
or depots, namely the subcutaneous and intraabdominal, each with its own characteristic metabolic, endocrine, paracrine, and autocrine
function.24 BAT can be found in the newborns and
neonates around many major vessels, the kidneys,
and in adrenal glands.25 In small mammals, such as
rodents, BAT persists throughout life. In larger
mammals and humans, BAT depots undergo a morphologic transformation in which they rapidly accumulate fat, become unilocular, and lose the
ultrastructural and molecular properties that define
them, including mitochondria.26,27 Because of this,
there are very few, if any, collections of BAT in adult
humans.
Although WAT and BAT distribution patterns
have been well documented, their morphology
has been defined on the basis of histologic features
alone, which are not sufficient to differentiate
between the two. Brown adipocytes may appear
white when not stimulated.26 Likewise, the morphology of white adipocytes changes progressively
during fasting. They can become elongated and
multilocularor so-called slim. More advanced
techniques are designed to detect the presence of

Khan et al 367

J AM ACAD DERMATOL
VOLUME 62, NUMBER 3

uncoupling protein-1 (UCP-1), a protein that is

unique to brown adipocytes. UCP-1 mRNA in adipose tissues is now considered a more accurate
method for identifying activated brown adipocytes.
Recent polymerase chain reaction (PCR) studies to
detect UCP-1 mRNA in adipose tissue from rodents
and humans have revealed the existence of scattered BAT within the WAT depots.28-30 Furthermore,
transdifferentiation of white adipocytes into brown
can be induced under certain conditions, thereby
refuting the notion that stem cell commitment to and
differentiation into the white or brown cell lineage is
permanent.26,31
Although WAT and BAT both express many of the
same adipocyte-specific genes needed for lipid synthesis and hydrolysisin addition to secreting hormones that regulate energy homeostasis, such as
leptin32BAT has emerged as an independent organ
with specific expression of proteins and unique
functions. Mitochondrial protein UCP-1 or thermogenin, which is expressed exclusively in BAT, is
responsible for mediating the basic function of
brown fat cellsnamely, the transfer of energy
from nutrition to heat. BAT also functions as a
protective factor against obesity.33,34
BAT activity is regulated by adrenergic sympathetic nervous system activity, which is governed by
selective regions in the hypothalamus.35,36 A thermogenic signal is transmitted by norepinephrine,
which in turn stimulates a-1 AR and b-AR, particularly b3-AR,30 thereby triggering lipolysis, brown
adipocyte proliferation, mitochondriogenesis, and
increased expression of UCP-1. Other positive regulators of UCP-1 include retinoic acid and thyroid
hormone T3.37,38 Even though the physiologic significance of these findings is still being elucidated, it
would be beneficial to understand the composition
of the adipose tissue of cellulite (BAT vs WAT) using
newer techniques, such as PCR studies. This may
provide a better understanding of the pathophysiologic mechanisms that might play a role in the
formation, persistence, and recurrence of celluliteeven after various therapeutic measures have
been employedin order to provide satisfactory
results. It may be inferred from the work of Berman
et al20 that the adipocytes of the gluteofemoral area
are white as opposed to brown based on their
responsiveness to catecholamines; they are predominantly a-2 ARepositive and therefore express an
antilipolytic effect as opposed to abdominal adipocytes, which are predominantly b-ARepositive.
There remain many unsolved mysteries that are
worth resolving regarding the terra incognita of
cellulite. Additional research needs to be done in
order to explore these unanswered questions.

MINOR FACTORS
Key points
d

It is controversial whether localized tissue

vascularity or inflammation plays a major
role in the etiology of cellulite
Cellulite can affect individuals regardless of
their body mass index

Cellulite tissue vascularity

It has been theorized that the process of cellulite
originates with deterioration of the dermal vasculature, particularly in response to altered precapillary
arteriolar sphincters in affected areas. This is coupled
with the deposition of hyperpolymerized glycosoaminoglycans (GAGs) in the dermal capillary walls
and within the ground substance between collagen
and elastin bundles. GAGs have hydrophilic properties, which leads to elevated excessive fluid retention in the dermis, adipocytes, and interlobular
septae. Edema can also lead to vascular compression, hypoxia, and capillary neoformation, resulting
in microhemorrages that are noted upon histologic
evaluation.12,39 Although Rossi and Vergnanini12 and
Curri39 have supported the findings of increased
edema and abundant GAG deposition at the lower
dermal/subcutaneous junction in patients affected
with cellulite, this observation has not been supported by others.7,9,11
Cellulite and postinflammatory changes
Based on the reported subjective tenderness upon
deep compression of the cellulite-affected areas,
inflammation has also been proposed as a possible
basis for its pathophysiology.40-42 Kligman41 has
reported the presence of macrophages and lymphocytes in the fibrous septae of cellulite biopsies, which
might be the cause of low grade inflammation that
results in dermal atrophy. Others, however, found no
evidence of inflammation in patients with
cellulite.1,7,9
Effects of body mass index on cellulite
Cellulite can affect individuals with higher body
mass index (BMI) measurements and those with
lower BMIs. Mirrashed et al42 compared the clinical
grades of cellulite with magnetic resonance imaging
(MRI) scans among individuals with different BMIs
and found a somewhat positive correlation (Fig 6).
They reported that affected individuals with higher
BMIs have a weaker, less dense connective tissue
structure, leading to increased extrusion of adipose
tissue lobules through the hypodermis. In this group,
the septae are primarily composed of collagen and
elastin fibers. The amount of extrusion was found to
be higher and the thickness of the dermis was found

368 Khan et al

J AM ACAD DERMATOL
MARCH 2010

Fig 6. Examples of sagittal magnetic resonance imaging scans of thigh skin. A and B,
Unaffected males with higher body mass index values. C and D, Cellulite-affected females with
higher body mass index values. Of note is the fact that a greater number of septae are oriented
rather perpendicular to the skin surface in obese women with cellulite. (Reprinted with
permission.42)

to be significantly lower. In affected individuals with

lower BMIs, they showed differences in the thickness
of the adipose tissue layer, with a significantly thicker
adipose layer in individuals with clinically evident
cellulite. No significant differences in dermal thickness were found in affected versus unaffected subjects with lower BMIs.
Cellulite grading based on clinical severity
Cellulite can be divided into three main grades
based on the clinical severity. Grade I is characterized by smooth skin without any dimpling upon
standing and laying down, but the skin adopts a
mattress-like configuration upon pinching, which
forces the fat into the reticular and papillary dermis.
In grade II cellulite, a mattress-like appearance of
cellulite is present upon standing but disappears
with the patient in the supine position. Grade III
cellulite can be seen in individuals who exhibit skin
dimpling upon standing and when they are in the
supine position; this can be exacerbated by pinching
the skin.42

CELLULITE AS VIEWED BY NONINVASIVE

IMAGING TECHNIQUES
Key points
d

Individuals affected with cellulite have herniations of adipose tissue into the dermis
Affected women have fewer septae arranged
rather perpendicularly at hypodermal level

as opposed to the criss-cross pattern seen in

unaffected men and women
The current concept of the anatomy of adipose
tissue derives from the histologic studies of
Nurnberger and Muller,7 who analyzed adipose
samples of unaffected men, unaffected women,
and women with cellulite. They reported deep
adipose indentations into the dermis in women,
but not in men. They also described the architecture
of fibrous septae, mainly oriented perpendicular to
the skin surface in women as opposed to criss-cross
pattern in men. A microanatomic description of
cellulite was revisited by Pierard et al9 and confirmed
by Rosenbaum et al,8 who reported a rather undulating dermohypodermal interface in women with
cellulite. The research by Querleux et al11 provided a
better understanding of the adipose tissue in affected
and unaffected males and females as seen by in vivo
MRI scans and spectroscopy (Figs 1, 3, and 4). Their
work clearly revealed increased hypodermal thickness in women with cellulite compared to unaffected
women (P \ .01), but there were no sex-related
differences. Furthermore, when comparing unaffected men to women, they found a thicker lamellar
adipose layer (inner) than the areolar (outer) adipose layer (P = .0001). In their study, the threedimensional architecture of fibrous septae revealed
that Camper fascia is a thin plane structure that is
parallel to the surface of the skin. Vertical spetae
act as suspenders, in contrast to the straight planes

J AM ACAD DERMATOL

Khan et al 369

VOLUME 62, NUMBER 3

proposed by Nurnberger and Muller.7 Querleux

et al11 also reported a randomly organized pattern
of fibrous septae with presence of localized thickened septae. Their MRI findings did not confirm the
hypothesis of an increase in water content of subcutaneous adipose tissue in the case of cellulite, as
previously reported.42 Of note is the fact that their
measurements were strictly limited to the fat lobule. Whether the overall increased fluid volume in
cellulite is confined to the connective tissue or
adipose tissue or both needs to be investigated.
REFERENCES
1. Scherwitz C, Braun-Falco O. So-called cellulite. J Dermatol Surg
Oncol 1978;4:230-4.
2. Curri SB. Aspects morpho-histochimiques et biochimiques du
tissue adipeux dans la dermohypodermose cellulitique. J Med
Esth 1976;5:183-91.
3. Binazzi M, Grilli-Cicioloni E. A proposito della cosidetta cellulite
e della dermato-panniculopatia edemato fibrosclerotica. Ann
It Derm Clin Sper 1977;31:121-5.
4. Ciporkin H, Paschoal LH. Atualizaca terapeutica e fisiopatogenica da Lipodistrofia Ginoide (LGD) cellulite. Sao Paulo,
Brazil: Livraria Editora Santos; 1992.
5. Bray GA. Obesity: basic considerations and clinical approaches.
Dis Mon 1989;35:449-537.
6. Lotti T, Ghersetich I, Grappone C, Dini G. Proteoglycans in socalled cellulite. Int J Dermatol 1990;29:272-4.
7. Nurnberger F, Muller G. So-called cellulite: an invented
disease. J Dermatol Surg Oncol 1978;4:221-9.
8. Rosenbaum M, Prieto V, Rudolph L, Ship A. An exploratory
investigation of the morphology and biochemistry of cellulite.
Plast Reconstr Surg 1998;101:1934-9.
9. Pierard GE, Nizet JL, Pierard-Franchimont C. Cellulite: from
standing fat herniation to hypodermal stretch marks. Am J
Dermatopathol 2000;22:34-7.
10. Agache P, Ovide MT, Laurent R. Mechanical factors in striae
distensae. In: Moretti GR, Rebora A, editors. Striae distensae.
Milan, Italy: Symposium Publication Division Brocades; 1976.
pp. 87-96.
11. Querleux B, Cornillon C, Jolivet O, Bittoun J. Anatomy and
physiology of subcutaneous adipose tissue by in vivo
magnetic resonance imaging and spectroscopy: relationships with sex and presence of cellulite. Skin Res Technol
2002;8:118-24.
12. Rossi AB, Vergnanini AL. Cellulite: a review. J Eur Acad
Dermatol Venereol 2000;14:251-62.
13. Krotkiewski M, Bjorntorp P, Sjostrom L, Smith U. Impact of
obesity on metabolism in men and women. Importance of
regional tissue distribution. J Clin Invest 1983;72:1150-62.
14. Roncari DA, Van RL. Promotion of human adipocyte precursor
replication by 17beta-estradiol in culture. J Clin Invest 1978;62:
503-8.
15. Berlan M, Galitzky J, Lafontan M. Heterogeneite fonctionnelle
du tissue adipeux: recepteurs adrenergiques et lipomobilisation. L Med Esth Et Chir Derm 1992;19:7-15.
16. Wahrenberg H, Lonnqvist F, Arner P. Mechanisms underlying
regional differences in lipolysis in human adipose tissue. J Clin
Invest 1989;84:458-67.
17. Leibel RL, Hirsch J. Site- and sex-related differences in
adrenoreceptor status of human adipose tissue. J Clin
Endocrinol Metab 1987;64:1205-10.

18. Leibel RL, Edens NK, Fried SK. Physiologic basis for the control of
body fat distribution in humans. Annu Rev Nutr 1989;9:417-43.
19. Mauriege P, Galitzky J, Berlan M, Lafontan M. Heterogenous
distribution of beta and alpha-2 adrenorecptor binding sites in
human fat cells from various fat deposits: functional consequences. Eur J Clin Invest 1987;17:156-65.
20. Berman DM, Nicklas BJ, Rogus EM, Dennis KE, Goldberg AP.
Regional differences in adrenoreceptor binding and fat cell
lipolysis in obese, postmenopausal women. Metabolism 1998;
47:467-73.
21. Ley CJ, Lees B, Stevenson JC. Sex- and menopause-associated
changes in body-fat distribution. Am J Clin Nutr 1992;55:950-4.
22. Cinti S. Adipocyte differentiation and transdifferentiation: plasticity of the adipose organ. J Endocrinol Invest 2002;25:823-35.
23. Fawcett DW. A comparison of histological organization and
cytochemical reactions of brown and white adipose tissues. J
Morphol 1952;90:363-405.
24. Smith SR, Lovejoy JC, Greenway F, Ryan D, deJonge L, de la
Bretonne J, et al. Contributions of total body abdominal fat,
subcutaneous adipose tissue compartments, and visceral
adipose tissue to the metabolic complications of obesity.
Metabolism 2001;50:425-35.
25. Houstek J, Vzek K, Pavelka S, Kopecky J, Krejcova E,
Hermanska J, et al. Type II iodothyronine 59-deiodinase and
uncoupling protein in brown adipose tissue of human newborns. J Clin Endorinol Metab 1993;77 328-7.
26. Himms-Hagen J. Does brown adipose tissue have a role in the
physiology or treatment of human obesity? Rev Endocr Metab
Disord 2001;2:395-401.
27. Penicaud L, Cousin B, Leloup C, Lorsignol A, Casteilla L. The
autonomic nervous system, adipose tissue plasticity, and
energy balance. Nutrition 2000;16:903-8.
28. Champigny O, Ricquier D. Evidence from in vitro differentiating cells that adrenoreceptor agonists can increase uncoupling protein mRNA level in adipocytes of adult humans: an
RT-PCR study. J Lipid Res 1996;37:1907-14.
29. Oberkofler H, Dallinger G, Liu YM, Hell E, Krempler F, Patsch W.
Uncoupling protein gene: quantification of expression levels
in adipose tissues of obese and non-obese humans. J Lipid Res
1997;38:2125-33.
30. Krief S, Lonnqvist F, Raimbault S, Baude B, Van Spronsen A,
Arner P, et al. Tissue distribution of b3-adrenergic receptor
mRNA in man. J Clin Invest 1993;91:344-9.
31. Tiraby C, Tavernier G, Lefort C, Larrouy D, Bouillaud F, Ricquier
D, et al. Acquirement of brown fat cell features by human
white adipocytes. J Biol Chem 2003;278:33370-6.
32. Cinti S, Frederich RC, Zingaretti MC, De Matteis R, Flier JS,
Lowell BB. Immunohistochemical localization of leptin and
uncoupling protein in white and brown adipose tissue.
Endocrinology 1997;138:797-804.
33. Cannon B, Nedergaard J. Brown adipose tissue: function and
physiological significance. Physiol Rev 2004;84:277-359.
34. Sell H, Deshaies Y, Richard D. The brown adipocyte: update on
its metabolic role. Int J Biochem Cell Biol 2004;36:2098-104.
35. Lowell BB, Spiegelman BM. Towards a molecular understanding of adaptive thermogenesis. Nature 2000;404:652-60.
36. Avram AS, Avram MM, James WD. Subcutaneous fat in normal
and diseased states: 2. Anatomy and physiology of white and
brown adipose tissue. J Am Acad Dermatol 2005;53:671-83.
37. Rial E, Gonzalez-Barroso M, Fleury C, Iturrizaga S, Sanchis D,
Jimenez-Jimenez J, et al. Retinoids activate proton transport
by the uncoupling proteins UCP1 and UCP2. EMBO J 1999;18:
5827-33.
38. Mostyn A, Pearce S, Stephenson T, Symonds ME. Hormonal
and nutritional regulation of adipose tissue mitochondrial

370 Khan et al

J AM ACAD DERMATOL
MARCH 2010

development and function in the newborn. Exp Clin Endocrinol Diabetes 2004;112:2-9.
39. Curri SB. Cellulite and fatty tissue microcirculation. J Cosmet
Toilet 1993;108:51-8.
40. Draelos Z, Marenus KD. Cellulite. Etiology and purported
treatment. Dermatol Surg 1997;23:1177-81.

41. Kligman AM. Cellulite: facts and fiction. J Geriatr Dermatol

1997;5:136-9.
42. Mirrashed F, Sharp JC, Krause J, Morgan J, Tomanek B. Pilot
study of dermal and subcutaneous fat structures by MRI in
individuals who differ in gender, BMI, and cellulite grading.
Skin Res Technol 2004;10:161-8.