J. Cosmet. Sci.) 56, 105-120 (March/April 2005) Quantitative model of cellulite: Three-dimensional skin surface topography, biophysical characterization, and relationship to human perception LOLA K. SMALLS, CAROLINE Y. LEE, JENNIFER WHITESTONE, W. JOHN KITZMILLER, R. RANDALL WICKETT, and MARTY 0. VISSCHER, The Skin Sciences Institute, Cincinnati Children's Hospital Research Foundation, Cincinnati, OH (L.K.S., C. Y.L., W.J.K. ) R.R. W., M.O. V.) Colleges of Pharmacy (L.K.S.) R.R. W.) and Medicine (W .j.K.), University of Cincinnati, Cincinnati, OH 54267 and Total Contact, Inc., Germantown, OH 45327 U. W.). Accepted for publication February 16, 2005. Presented in part at the 23rd Conference of the International Federation of the Societies of Cosmetic Chemists, Orlando) Florida, October 24-27, 2004, and at the U.S. Regional Meeting of the International Society for Bioengineering and the Skin, Orlando, Florida, October 28-30, 2004. Synopsis Gynoid lipodystrophy (cellulite) is the irregular, dimpled skin surface of the thighs, abdomen, and buttocks in 85% of post-adolescent women. The distinctive surface morphology is believed to result when subcu­ taneous adipose tissue protrudes into the lower reticular dermis, thereby creating irregularities at the surface. The biomechanical properties of epidermal and dermal tissue may also influence severity. Cellulite-affected thigh sites were measured in 51 females with varying degrees of cellulite, in 11 non-cellulite controls, and in 10 male controls. A non-contact high-resolution three-dimensional laser surface scanner was used to quantify the skin surface morphology and determine specific roughness values. The scans were evaluated by experts and nai"ve judges (n = 62). Body composition was evaluated via dual-energy x-ray absorptiometry dermal thickness and the dermal-subcutaneous junction were evaluated via high-resolution 3D ultrasound and surface photography under compression. Biomechanical properties were also measured. The roughness parameters Svm (mean depth of the lowest valleys) and Sdr (ratio between the roughness surface area and the area of the xy plane) were highly correlated to the expert image grades and, therefore, designated as the quantitative measures of cellulite severity. The strength of the correlations among na·ive grades, expert grades, and roughness values confirmed that the data quantitatively evaluate the human perception of cellulite. Cellulite severity was correlated to BMI, thigh circumference, percent thigh fat, architecture of the dermal-subcutaneous border (ultrasound surface area, red-band SD from compressed images), compliance, and stiffness (negative correlation). Cellulite severity was predicted by the percent fat and the area of the dermal-subcutaneous border. The biomechanical properties did not significantly contribute to the predic­ tion. Comparison of the parameters for females and males further suggest that percent thigh fat and surface area roughness deviation are the distinguishing features of cellulite. 105

106 JOURNAL OF COSMETIC SCIENCE INTRODUCTION Gynoid lipodystrophy (cellulite), the unattractive cottage cheese-like dimpling of the thighs, abdomen, and buttocks, affects 85% of post-adolescent women (1,2). Cellulite treatment is a high priority for the pharmaceutical and cosmetic industries (2-10). Products (11-15), supplements (16), and massage techniques (8,9) purport to treat cellulite, presumably by reducing the appearance of the dimpled, lumpy skin. The uneven skin surface texture is attributed to the three-dimensional (3D) architecture of the hypodermal connective tissue (14, 17-20). In females, fat cell chambers, "papillae adiposae," are sequestered by connective tissue septa, positioned in a radial and arched manner and anchoring the dermis to the muscle fascia. The subcutaneous fat cell chambers bulge into the dermis, thereby changing the skin surface appearance (13). The literature on the etiology of cellulite and the effectiveness of treatments to ameliorate the condition is limited (4,6,14,16,21), given the prevalence. The surface features of cellulite are believed to result when subcutaneous adipose tissue protrudes into the lower reticular dermis, thereby creating irregularities at the epidermal surface. The biomechanical properties of the epidermis and dermis may also influence the severity. Cellulite is not specific to overweight females, but added weight may cause enlargement of the fat lobules, further protrusion into the dermis, and exacerbation of the condition (2, 17). Weight loss is reported to diminish cellulite, but it may not alter the underlying dermal-subcutaneous structures (1 7, 19). Identification of key factors responsible for the visual appearance of cellulite will help to facilitate the development and selection of effective treatments. We conducted a set of noninvasive biophysical measurements of the cellulite-affected tissue and determined the specific factors that contribute to cellulite severity. The surface morphology was quantified with a non-contact three-dimensional laser scanning system to generate surface roughness parameters and provide a standardized measure of severity. In the literature, cellulite severity is generally evaluated with various visual and photographic methods, although accepted standards have not yet emerged. We related the technical measures of severity to nai·ve and subject assessment using a 0-9 category scale. Quantitative, reproducible methods will facilitate effective comparison of treat­ ments across studies. Furthermore, treatment effectiveness will be judged by the patient/ consumer based on the impact on cellulite severity and appearance. Ultimately, evalu­ ation methods must be linked to human perception of severity and change. MATERIALS AND METHODS SUBJECTS Fifty-one females with visible cellulite were recruited from several weight-loss programs (medication, liquid diet, Weight Watchers®, and bariatric surgery). Eleven females without visible cellulite were controls. Individuals who were pregnant, had an active skin condition (e.g., rash, wound) on the thigh, or had been treated for cellulite within three months were excluded from participation. The Institutional Review Boards of Cincinnati Children's Hospital Medical Center and the University of Cincinnati ap­ proved the research protocol. All subjects provided written informed consent for par­ ticipation.