Journal Logo

Cosmetic: Original Articles

The Anatomical Basis of Cellulite Dimple Formation: An Ultrasound-Based Examination

Whipple, Lauren A. M.D.; Fournier, Craig T. M.D.; Heiman, Adee J. M.D.; Awad, Amanda A. B.S.; Roth, Malcolm Z. M.D.; Cotofana, Sebastian M.D., Ph.D.; Ricci, Joseph A. M.D.

Author Information
Plastic and Reconstructive Surgery: September 2021 - Volume 148 - Issue 3 - p 375e-381e
doi: 10.1097/PRS.0000000000008218
  • Free
  • SDC
  • Associated Video
  • Watch Video
  • Press Release


Cellulite is a common aesthetic condition that affects the vast majority of women, and is characterized by the inhomogeneous appearance of skin overlying the gluteal and posterior thigh regions.1 This change in quality of the skin and surface irregularities may result in an appearance of “dimpling” or indentations of a region that can be a source of great distress and embarrassment for a woman. This complex condition is multifactorial and influenced by an array of factors, including fluctuations in weight, hormone activity, sex, and age.2–4 Despite its high prevalence (80 to 90 percent) in postpubertal female patients of various ethnic groups, little is known regarding the anatomical basis of cellulite; as a result, there exists a wide array of suboptimal treatment options.1

Previous studies have investigated the pathophysiology of cellulite and have found that hormonal expression may be an important factor in the change of subcutaneous tissue. A study by Emanuele et al. found that adiponectin, a hormone produced in the subcutaneous fat with vasodilatory effects, is down-regulated in tissue affected by cellulite.3 In addition to changes in vascularity; other studies have examined how patient-controlled factors such as dramatic weight loss may impact cellulite deposition. A study by Smalls et al. quantitatively examined cellulite in women who underwent a supervised weight loss program and found that weight loss alone may not improve the appearance of cellulite. Although the majority of patients did find improvement with weight loss, a group of subjects paradoxically had a worsening effect of dimpling with drastic weight loss.4

As the cause of cellulite remains unclear, recent research into this topic has shifted from prevention of cellulite toward its treatment. Previously proposed treatments have included topical treatments with caffeine and retinoids, laser and radiofrequency devices, and cryolipolysis.5,6 One treatment option that has been gaining interest in the literature recently is the tissue stabilized-guided subcision approach or “subcision approach,” where a needle is inserted beneath the skin into the area of subcutaneous depression to release any potential tethering of the skin to the underlying fascia, postulated to cause dimpling.7 Several newly developed devices that use the tissue stabilized subcision technique have been associated with high levels of patient satisfaction and give promising results for a treatment option.8 Despite this array of treatments modalities, there has been little prior research regarding the anatomical basis of cellulite to support or refute such interventions.2 A 2002 study by Querleux et al. used magnetic resonance imaging to demonstrate that areas of cellulite have a higher percentage of fibrous septa in the subcutaneous tissue, perhaps contributing to the indentations appreciated as cellulite dimples.9 Unfortunately, this study was unable to determine the precise orientation of these septa (perpendicular or oblique to skin) or where in the subcutaneous tissue these bands originated. Further studies, also using magnetic resonance imaging, have demonstrated improvement in these fascial bands after treatment with a subcision approach.10

Unfortunately, although magnetic resonance imaging is effective at demonstrating subcision release, this imaging modality is expensive and impractical for physicians to use with the subcision device in real time. The ability to visualize the release of fascial bands may not only aid the physician performing such a procedure, but may also help to explain and support treatment results to the patient. Ultrasound represents an inexpensive device, readily available in most offices, that not only may aid in further defining and visualizing the subcutaneous composition of cellulite, but may further elucidate whether treatment options such as the subcision approach are warranted. To date, no study has evaluated the utility of ultrasound in the identification of the fibrous septa causing cellulite and in delineating their anatomical orientation. This study aims to demonstrate whether the cellulite-associated fascial bands are easily visualized by ultrasound, thus proposing an accessible imaging option to perform in real time during the subcision technique.


Patient Selection

Between February of 2019 and April of 2020, 50 consecutive female volunteers were examined for the presence of cellulite. Patients were considered for voluntary participation in this study after meeting several criteria: age range from 20 to 55 years; healthy without contributing medical comorbidities; body mass index less than 40 kg/m2; and presentation to the clinic for a reason other than cellulite, liposuction, or body contouring. Body mass index from 18.5 to 24.9 kg/m2 was considered normal weight; body mass index of 25 to 29.9 kg/m2 was considered overweight, and body mass index greater than 30 kg/m2 was considered obese. All patients provided written consent and photographic consent, and all participated voluntarily without compensation before inclusion in the study, in accordance with the institutional review board guidelines at our institution. After agreeing to participate, cellulite grading and dimple analysis was performed by two authors.

Cellulite Dimple Analysis

The paired right and left posterior thigh and buttocks of each patient were visually inspected, and the most severe cellulite dimples were selected. For the purposes of analysis, each area was considered separately (buttock and thigh) and the contralateral regions considered independently. Accordingly, each of the 50 patients had four sites evaluated for a total of 200 anatomical sites. If no cellulite dimple was present in the areas examined, no further ultrasound analysis of that anatomical area was performed. Each dimple was graded with the Hexsel Cellulite Severity Scale, which takes into account the number, extent, and severity of the dimple.11 The Hexsel Cellulite Severity Scale is a validated and objective scale that grades the severity of the dimples on a scale of 1 to 15 (where 1 to 5 = mild, 6 to 10 = moderate, and 11 to 15 = severe). The total score is composed of the sum of the following five items graded from 0 to 3: number of cellulite depressions, depth of depressions, morphologic appearance of skin, laxity of skin, and the Nürenberger and Müller numeric grade.11 The Hexsel Cellulite Severity Scale pictorial reference was used as a control for scoring each dimple.11 (See Figure, Supplemental Digital Content 1, which shows the Hexsel Cellulite Severity Scale, used to grade each cellulite dimple. Reproduced with permission from Wiley & Sons Publishers, Each cellulite dimple was also graded on a simplified cellulite severity grade of 0 to 3, also known as the most severe dimple scale score, with 0 being no dimpling, 1 being minimal, 2 being moderate, and 3 characterizing severe dimpling.11

Ultrasound Measurements and Analysis

After visual analysis, an ultrasound probe with a 15-MHz transducer was applied to the preselected dimples to identify any subcutaneous related anatomical structures such as fibrous bands or blood vessels. First, the skin depth at each dimple was measured to further characterize the morphology of skin at the level of the cellulite. This measurement was from the epidermis to dermis and was clearly delineated on the ultrasound image. (See Figure, Supplemental Digital Content 2, which shows representative measurements from ultrasound examination of cellulite dimples. Yellow arrow, skin thickness; green arrow, skin to superficial fascia thickness; orange arrow, skin to deep fascia thickness; red arrowheads, superficial fascial layer; blue arrowheads, deep fascial layer, Next, measurements were made from the skin to the superficial fascia and skin to the deep fascia to identify the thickness of the superficial and deep subcutaneous tissues at the level of the dimple, respectively. The superficial fascia, identified as a hyperechoic area within the subcutaneous tissue, is analogous to the Scarpa fascia over the abdomen. The deep fascia lies anatomically just superficial to the muscles.

After measurements were taken, each cellulite dimple was examined on ultrasound for an associated fascial band extending from the skin through the subcutaneous tissue toward the fascia. If a fascial band was present, it was further characterized by its orientation as either straight (perpendicular in orientation to the skin) or oblique (at a nonperpendicular angle to the skin). Furthermore, the fascial band was further defined as originating from either the superficial fascia or the deep fascia (Fig. 1). [See Video (online), which demonstrates the identification of an oblique subcutaneous fascial band during the clinical evaluation of cellulite dimples.] The ultrasound was then switched to a color Doppler mode and each dimple was examined for the presence of an associated blood vessel. All data were collected and analyzed using Microsoft Excel (Microsoft Corp., Redmond, Wash.).

{"href":"Single Video Player","role":"media-player-id","content-type":"play-in-place","position":"float","orientation":"portrait","label":"Video.","caption":"This video demonstrates the identification of an oblique subcutaneous fascial band during the clinical evaluation of cellulite dimples.","object-id":[{"pub-id-type":"doi","id":""},{"pub-id-type":"other","content-type":"media-stream-id","id":"1_axum625a"},{"pub-id-type":"other","content-type":"media-source","id":"Kaltura"}]}

Fig. 1.
Fig. 1.:
An oblique fascial band as seen on ultrasound examination of cellulite dimple. The fascial band is highlighted with a yellow circle.


A total of 50 female patients were included in the study, with a potential 200 anatomical sites for examination. Of these, 173 sites (86.5 percent) had evidence of clinically identifiable cellulite dimples, and the remaining 27 sites (13.5 percent) were excluded from ultrasound examination. Of these 173 sites, 81 (46.8 percent) were on the thigh and 92 were on the buttock (53.2 percent). Patients presented to the office for a variety of different reasons; none presented for consultation related to body contouring, liposuction, or cellulite. Patient age ranged from 21 to 55 years, with an average of 36 years; the average body mass index was 25.7 kg/m2. Recorded Hexsel Cellulite Severity Scale grades for both the buttock and thigh cellulite dimples ranged from 1 to 15. On evaluation of each buttock dimple, the average Hexsel Cellulite Severity Scale score was 7.12, with a most severe dimple scale score of 1.78. In comparison, the average Hexsel Cellulite Severity Scale thigh grade was 7.69, with a most severe dimple scale score of 1.87. These values were not statistically different from each other with regard to anatomical location of cellulite dimples (Table 1).

Table 1. - Overall Cellulite Severity Scores, by Anatomical Location of Cellulite Dimples
Buttock Site Thigh Site p
Mean HCSS score 7.12 7.69 NS
Mean MSD scale score 1.78 1.87 NS
HCSS, Hexsel Cellulite Severity Scale; MSD, most severe dimple; NS, not significant.

By ultrasound examination, skin thickness was measured at the exact location of each dimple selected. The average skin thickness was 0.28 cm for the buttock dimples and 0.26 cm for the thigh (p value not statistically different). The mean distances from the skin to the superficial fascia and the skin to the deep fascia were not statistically different between the buttock dimples and thigh dimples (0.84 cm versus 0.82 cm, and 1.60 cm versus 1.62 cm, respectively) (Table 2). Of the 173 cellulite dimples examined on ultrasound, 169 (97.68 percent) of the dimples had an associated fascial band at the area of cellulite indentation, whereas four (2.32 percent) of the dimples did not have a fascial band present on ultrasound (Table 2). Cellulite dimple bands, if present, were further characterized into having either an oblique or straight orientation through the subcutaneous tissues. Of the fascial bands present, the majority (92.31 percent) had an oblique orientation compared to a straight arrangement (7.69 percent) (Table 2). The obliquely oriented fascial band was the dominant variant seen in both the buttock (83.7 percent) and the thigh (85.2 percent) sample sites. There was no statistical difference in the band orientation between the two anatomical sites. In addition, in both the thigh and the buttock regions, the majority of bands were seen to clearly be originating from the superficial fascia. The buttock dimples had a slightly higher percentage of bands originating form the superficial facia (93.4 percent) compared to the thigh region (86.4 percent), but this was not statistically significant. The origin of the fascial band was then characterized as coming from either the superficial or deep fascia. Of the 169 fascial bands examined in both the buttock and the posterior thigh areas, the majority of dimples did not have a vessel present [n = 152 (89.02 percent)]. In comparison, 17 cellulite dimples did have a vessel visualized with color Doppler flow mode (10.98 percent) (Table 2). There was no statistical difference with regard to blood vessel presence or absence between the two anatomical sites (Table 2).

Table 2. - Summary of Ultrasound Findings Associated with Cellulite Dimples, by Anatomical Location
Buttock Site (%) Thigh Site (%) p
No. of dimples 92 81
Average skin thickness, cm 0.28 0.26 NS
Depth of superficial fascia, cm 0.84 0.82 NS
Depth of deep fascia, cm 1.60 1.62 NS
Fascial band
 Present 90 (97.8) 79 (97.5) NS
 Absent 2 (2.2) 2 (2.5)
Band orientation
 Oblique 77 (83.7) 69 (85.2) NS
 Vertical 13 (14.1) 10 (12.3)
 Absent 2 (2.2) 2 (2.5)
Band origin
 Superficial fascia 86 (93.4) 70 (86.4) NS
 Deep fascia 4 (4.4) 9 (11.1)
 Absent 2 (2.2) 2 (2.5)
Blood vessel
 Absent 81 (90.0) 71 (89.9) NS
 Present 9 (10.0) 8 (10.1)
NS, not significant.

When stratified by body mass index, analysis of variance revealed a statistically significant increase in Hexsel Cellulite Severity Scale and most severe dimple scale scores in the overweight and obese populations compared to the normal body mass index population (Fig. 2). Overweight and obese patients also demonstrated a statistically significant increase in the likelihood of having a blood vessel present in their fibrous band when seen on ultrasound, compared to normal weight individuals (p = 0.01). Band characteristics (band origination, band orientation, or presence of vessel) did not vary significantly when stratified by patient age or cellulite severity scores (both most severe dimple scale and Hexsel Cellulite Severity Scale scores).

Fig. 2.
Fig. 2.:
Cellulite severity scores, stratified by body mass index. CSS, cellulite severity score; MSD, most severe dimple scale; BMI, body mass index.


Cellulite is a highly prevalent, aesthetically disturbing condition characterized by the inhomogeneous appearance of skin without predilection for race, weight, or body type.1 Multiple causes are thought to contribute to cellulite formation, including vascular, inflammatory, hormonal, and structural causes, but little is known about the anatomy of cellulite.12,13 Several proposed treatments, especially minimally invasive options, have been used with limited success likely because of the uncertainty regarding the underlying anatomical structural causes of cellulite.14 Despite this gap in knowledge, various treatment options continue to be used with varying efficacy because of the lack of comprehensive data regarding patient-specific anatomy and cellulite morphology.13 This study aimed to demonstrate the use of ultrasound as a reliable, point-of-care tool for the anatomical analysis of cellulite to aid in the understanding of its subcutaneous composition, which can be used to further the development and evaluation of treatment options.

The utility of ultrasound to image cellulite was evaluated by identifying and assessing cellulite dimples and the underlying soft tissues in comparison with a validated cellulite severity scale. The majority of cellulite dimples evaluated in this study had an underlying fibrous band identified sonographically, a finding consistent with previous descriptions of cellulite using magnetic resonance imaging, suggesting that ultrasound is a valid technique to image cellulite within a wide range of ages, body mass indexes, and cellulite severity scale grades.10,15 Magnetic resonance imaging is recognized as a reliable technique to visualize subcutaneous structures involved in cellulite, but previous studies using magnetic resonance imaging have not yet described the comprehensive anatomy of the associated subcutaneous structures, such as the fibrous band orientation, origination, and associated blood vessels.15 In addition, magnetic resonance imaging is costly, takes time for the examination to be scheduled, and it is not readily available at all centers, making it extremely impractical for real-time imaging during procedures to treat cellulite.

This study demonstrated a preponderance of obliquely oriented fibrous bands that originated from the superficial fascial layer of the subcutaneous tissues, which has previously not been reported in the literature to date. In addition, and perhaps equally importantly, the ultrasound examination reported approximately 10 percent of bands to contain an associated blood vessel. Interestingly, these findings were not statistically different when comparing the buttock and thigh locations, which are two of the most common anatomical locations to find cellulite. These characteristics, and their consistency across anatomical sites, have a significant impact on the development and efficacy of proposed treatments for cellulite, and the complication risk. Several minimally invasive treatments specifically target the release of these fibrous septa contributing to cellulite dimples; however, until this point, it was unclear precisely where in the subcutaneous tissue these bands existed so they could be released accurately and precisely.16 These results provide insight into the clinical observation that the depth of subcision is a major determining factor for outcomes.5 Because the majority of facial bands originate from the superficial fascia, it is possible that a surgeon using a depth below the superficial fascial system might fail to lyse the causative band, leading to disappointing results (Table 2). In addition, given that the fascial bands are most commonly obliquely oriented away from the dimple, surgeons might need to release the subcutaneous tissue in a radius wider than previously considered. Given the traditionally high rates of recurrence of cellulite after treatment, it seems likely that real-time visualization of these fascial bands with an ultrasound device during release might be able to improve the outcomes and decrease recurrence rates.

Furthermore, understanding the real-time anatomy of cellulite dimples, as demonstrated in this study with ultrasound, can also help clinicians avoid complications associated with the subcision technique. Doppler evaluation of the area adjacent to the identified fibrous bands did not reveal any significant vascular structures in the majority of cases (89.02 percent). This supports a low (but not negligible) risk of bleeding complications following current subcision techniques.17–19 We believe that using ultrasound in real time, while performing a release of the subcutaneous tissues (with subcision or other technique), can help surgeons from causing avoidable bruising and bleeding in patients undergoing this procedure. This is an important consideration, as the subcision approach has gained popularity over the past several years.16,20–22 Despite promising results using the subcision method, the outcomes of these various treatment modalities are often not reproducible across practitioners, an issue that real-time imaging of subcutaneous anatomy may address. Understanding the subcutaneous anatomy—including the orientation, origin, and presence of blood vessels in relation to these bands—could improve efficacy and longevity of results, decrease hematoma rates, and limit pain and discomfort of excessive tissue dissection. The use of ultrasound to evaluate cellulite may lead to great advances in potential treatment options.

Some limitations of this study that must be acknowledged include a relatively small sample size given this pilot study, the user-dependent nature of the ultrasound examination of soft tissues, and the subjectivity of the photonumeric cellulite severity scale (despite its previous validation). We aimed to control for the latter by having two independent examiners score the cellulite dimples independently at the same office visit. Future prospective studies will be needed on a significantly greater number of patients to confirm the results of this study and to more closely examine the real-time application of ultrasound evaluation during subcision release of cellulite.


Cellulite has become a popular topic with many proposed treatment modalities, yet no detailed and comprehensive anatomical data on the subcutaneous arrangement of cellulite in relation to dimple formation were previously available. To the authors’ knowledge, this is the first study to demonstrate that ultrasound is a valid and effective technique to image cellulite. As the majority of cellulite dimples examined had an associated fascial band on ultrasound, we may then propose that ultrasound has additional utility to monitor cellulite treatment procedures in real time. The knowledge of fibrous band orientation and origination and the presence of blood vessels may improve current treatment options for cellulite reduction, specifically, treatment modalities aimed at fascial band lysis. In addition, these findings may be used in conjunction with ultrasound to guide treatment in real time, by avoiding blood vessels and performing the subcision technique at the correct depth, allowing surgeons to achieve the optimal treatment while avoiding complications.


1. Emanuele E. Cellulite: Advances in treatment. Facts and controversies. Clin Dermatol. 2013;31:725–730.
2. Rudolph C, Hladik C, Hamade H, et al. Structural gender dimorphism and the biomechanics of the gluteal subcutaneous tissue: Implications for the pathophysiology of cellulite. Plast Reconstr Surg. 2019;143:1077–1086.
3. Emanuele E, Minoretti P, Altabas K, Gaeta E, Altabas V. Adiponectin expression in subcutaneous adipose tissue is reduced in women with cellulite. Int J Dermatol. 2011;50:412–416.
4. Smalls LK, Hicks M, Passeretti D, et al. Effect of weight loss on cellulite: Gynoid lypodystrophy. Plast Reconstr Surg. 2006;118:510–516.
5. Friedmann DP, Vick GL, Mishra V. Cellulite: A review with a focus on subcision. Clin Cosmet Investig Dermatol. 2017;10:17–23.
6. Adjadj L, SidAhmed-Mezi M, Mondoloni M, Meningaud JP, Hersant B. Assessment of the efficacy of cryolipolysis on saddlebags: A prospective study of 53 patients. Plast Reconstr Surg. 2017;140:50–57.
7. Uebel CO, Piccinini PS, Martinelli A, Aguiar DF, Ramos RFM. Cellulite: A surgical treatment approach. Aesthet Surg J. 2018;38:1099–1114.
8. Kaminer MS, Coleman WP III, Weiss RA, Robinson DM, Grossman J. A multicenter pivotal study to evaluate tissue stabilized-guided subcision using the Cellfina device for the treatment of cellulite with 3-year follow-up. Dermatol Surg. 2017;43:1240–1248.
9. 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;8:118–124.
10. Hexsel D, Dal Forno T, Hexsel C, et al. Magnetic resonance imaging of cellulite depressed lesions successfully treated by subcision. Dermatol Surg. 2016;42:693–696.
11. Hexsel D, Dal’Forno T, Hexsel C, Rodrigues T. Cellulite severity scale. JAAD. 2007;56:AB59.
12. Sadick N. Treatment for cellulite. Int J Womens Dermatol. 2019;5:68–72.
13. Christman MP, Belkin D, Geronemus RG, Brauer JA. An anatomical approach to evaluating and treating cellulite. J Drugs Dermatol. 2017;16:58–61.
14. Nikolis A, Enright KM, Sapra S, Khanna J. A multicenter, retrospective evaluation of tissue stabilized-guided subcision in the management of cellulite. Aesthet Surg J. 2019;39:884–892.
15. Hexsel DM, Abreu M, Rodrigues TC, Soirefmann M, do Prado DZ, Gamboa MM. Side-by-side comparison of areas with and without cellulite depressions using magnetic resonance imaging. Dermatol Surg. 2009;35:1471–1477.
16. Kaminer MS, Coleman WP III, Weiss RA, Robinson DM, Coleman WP IV, Hornfeldt C. Multicenter pivotal study of vacuum-assisted precise tissue release for the treatment of cellulite. Dermatol Surg. 2015;41:336–347.
17. Amore R, Amuso D, Leonardi V, et al. Treatment of dimpling from cellulite. Plast Reconstr Surg Glob Open. 2018;6:e1771.
18. Zerini I, Sisti A, Cuomo R, et al. Cellulite treatment: A comprehensive literature review. J Cosmet Dermatol. 2015;14:224–240.
19. Conti G, Zingaretti N, Amuso D, et al. Proteomic and ultrastructural analysis of cellulite: New findings on an old topic. Int J Mol Sci. 2020;21:E2077.
20. DiBernardo BE, Sasak GH, Petti CA, Hunstad JP, Burns AJ, Katz BE. Multi center evaluation for the treatment of cellulite using a minimally invasive laser with side firing subdermal fiber. Plast Reconstr Surg. 2012;130(Suppl):29.
21. Pérez Atamoros FM, Alcalá Pérez D, Asz Sigall D, et al. Evidence-based treatment for gynoid lipodystrophy: A review of the recent literature. J Cosmet Dermatol. 2018;17:977–983.
22. Ibrahim O, Haimovic A, Lee N, Kaminer MS. Efficacy using a modified technique for tissue stabilized-guided subcision for the treatment of mild-to-moderate cellulite of the buttocks and thighs. Dermatol Surg. 2018;44:1272–1277.

Supplemental Digital Content

Copyright © 2021 by the American Society of Plastic Surgeons