JOURNAL OF COSMETIC SCIENCE 248 Alpha-hydroxy acids (AHAs) are a class of compounds commonly used for chemical peel- ing, and glycolic acid is the most extensively studied of these acids (6). According to these studies, glycolic acid appears to induce an acid-dependent discohesion of corneo- cytes, and when used for a long period of time in high concentration, it is able to increase cell proliferation of the basal epithelial cells in the epidermis, elastic fi bers, and collagen (7,8). Many commercial skin-care products containing glycolic acid are proposed to coun- teract photoaging, decrease acne and pigmentary changes, or reduce stretch marks (8,9). As reported in several studies, better response is obtained only when glycolic acid is used at higher concentrations (50–70%) (10,11), increasing the risk of skin irritation. The use of glycolic acids in chemical peeling is strictly correlated with some undesirable side effects such as persistent erythema and pruritus, burning, post-infl ammatory hyper/ hypopigmentation, hypertrophic scarring, and infectious complications (3,11,12). More- over, recent experimental studies demonstrate that short-term application of glycolic acid sensitizes the skin to the damaging effects of UV light (13). To improve the safety of products, committees set up by associations of cosmetics manufacturers in Europe and in the USA recommend similar guidelines and, in particular, pH values higher than 3.5 and alpha-hydroxy acids contents lower than 7–10% (14). Recently, in order to achieve a balance between performance and risks, many common organic acids and combinations of them, such as mandelic acid, lactic acid, and natural acids from fruits (such as tomatoes, lemons, grapefruits, oranges, and limes) have been used in commercial products. The aim of the present study was to compare the effectiveness and the safety of different AHAs (glycolic acid, mandelic acid, and a blend of organic acids from grape juice) in skin exfoliation by objective instrumental methods. To evaluate the effi cacy of the exfoliating agents, a new experimental in vivo protocol based on DHA (dihydroxyacetone)-induced skin pigmentation and a non-invasive instrumental method was used. DHA is a three- carbon sugar, formally a derivate of glycerol, and it is the most common and safe cosmetic ingredient used in sunless tanning products (15). The pigmentation produced by DHA is the result of the chemical reactions (Maillard reaction) between the DHA and the amino acids of the corneocytes in the upper layers of the stratum corneum, forming poly- meric colored substances called melanoidins (15–17). Since DHA is bound in an irrevers- ible way to the free amino groups, the resultant color lasts for several days on the skin and is only removed by natural skin renewal. Skin regeneration occurs by the continuous generation of new cells in the basal layer that rise through the epidermal layers of the skin until they reach the stratum corneum, where the skin cells die and eventually fall or slough off (18). Therefore, the color intensity is directly related to the amount of DHA bound in the skin, and the durability of the staining by DHA is strictly dependent on the rate of skin cell renewal. On the basis of this assumption, in this study we tried to demonstrate that the use of DHA-induced pigmentation could be a valid method to estimate the activity of exfoliat- ing agents in promoting skin regeneration. For an objective evaluation of DHA-induced pigmentation, the study was carried out by non-invasive instrumental refl ectance spec- trophotometry, and spectral data were used to quantify skin color intensity. The safety profi le of the AHAs versus skin was studied by two different in vivo studies: the evaluation of skin erythema induced by topical application of acids at different con- centrations (10%, 30%, and 50% w/w) and the increase in sensitivity to UV light exposure

EFFICACY AND TOLERANCE OF EXFOLIATING AGENTS 249 in cutaneous sites previously treated with AHAs. Both in vivo evaluations were monitored by refl ectance spectrophotometry. MATERIALS AND METHODS SUBJECTS In vivo experiments were performed on twenty healthy volunteers (females/males 14:6) of skin types II and III, aged 25–35 years. Between October 2006 and Septem- ber 2007, the volunteers were recruited after medical screening that included fi lling in a health questionnaire and physical examination of the application sites. Subjects exhibiting such features as sunburn, suntan, burn marks, or any other active lesions that might interfere with evaluation were excluded from the study. After they were fully informed of the nature of the study, substances, and procedures involved, the subjects gave their written consent. For the period of the studies, in vivo experiments were carried out on the volar forearms of each volunteer. Each subject rested for 15 minutes before the experiments, and room conditions were set at 22° ± 2°C and 40–50% relative humidity. Two research assistants were responsible for all recruit- ment and data collection. TEST MATERIAL Glycolic acid (70% cosmetic grade) and mandelic acid were supplied by A.C.E.F. (A.C.E.F. s.p.a., Fiorenzuola, Piacenza, Italy). Organic acids from white grape juice (a solution con- taining a blend of tartaric acid (18.5%), malic acid (12%), citric acid (3%), lactic acid (2.5%), gluconic acid (3.5%) and shikimic acid (0.5%)) were supplied by Bionap (Renegrape®, Bionap s.r.l., Italy). Exfoliating gel formulations were prepared by gelifi ca- tion with xanthan gum (1% w/w) of aqueous solutions containing three different concen- trations of glycolic acid (formulations labeled GLY), mandelic acid (formulations labeled MAN), and grape acids (formulations labeled GA). Standard samples of GLY, MAN, and GA were weighted and dissolved in water, obtaining solutions with a fi nal concentration of 10%, 30%, and 50% w/w for each acid. The pH of the test materials was adjusted to 3.5 by using sodium hydroxide. The test materials were alphabetically coded by the manufacturer, and neither the research assistants nor the volunteers knew the content of the blinded formulations. A sunless tanning formulation containing 5% w/w dihydroxy- acetone (A.C.E.F. s.p.a., Fiorenzuola, Piacenza, Italy) was prepared by mixing DHA in water and stirring for ten minutes. INSTRUMENTS Skin refl ectance spectra were recorded using a refl ectance visible spectrophotomer, X-Rite model 968 (XRite Inc. Grandville, MI), having 0° illumination and a 45° viewing angle, calibrated and controlled as previously reported (19). Refl ectance spectra were obtained over the wavelength range of 400–700 nm using illuminant C and a 2° standard ob- server.