JOURNAL OF COSMETIC SCIENCE 252 DATA ANALYSIS All data obtained were submitted to a statistical analysis. All statistical comparisons in instrumental assessment were evaluated using repeat-measure analysis of variance (ANOVA) followed by the Bonferroni-Dunn post-hoc pair-wise comparison procedure. A p value of less than 0.05 was considered signifi cant. RESULTS IN VIVO EVALUATION OF THE EXFOLIATING EFFECTS OF THE FORMULATIONS ON DHA-INDUCED SKIN PIGMENTATION The effects of 10% GLY, 10% MAN, or 10% GA formulations on skin exfoliation rates were evaluated by objective instrumental observation of in vivo DHA-induced skin pigmentation disappearance. DHA-induced skin pigmentation was determined as a melanin index by refl ectance spectrophotometry. For each subject, skin refl ectance spectra were recorded and melanin indices were calculated over the monitoring period of the study. The trends of skin pigmentation disappearance are reported in Figure 1 by plotting M.I values versus time (days). As can be seen, after DHA-pigmentation, the formulations showed different trends in M.I. value reductions over time. Moreover, to better compare the skin exfoliation rate, the “recovery time” (RT) values are given in Table I for each subject admitted into the study. The results showed that 10% GLY, 10% MAN, and 10% GA formulations were effective in inducing skin M.I. reduction in comparison with the non-treated site (CONTR) (p 0.05). An RT value of 10% GLY was signifi cantly lower than an RT value of 10% GA (p 0.05). Furthermore, a signifi cant difference was found between the RT value of 10% MAN and that of the 10% GA formulation (p 0.05). Figure 1. Trends of the melanin index (M.I.) vs time (days) for subjects recruited in the CONTR (no topical treatment), 10% GLY (glycolic acid), 10% MAN (mandelic acid), and 10% GA (grape acids) groups.

EFFICACY AND TOLERANCE OF EXFOLIATING AGENTS 253 IN VIVO EVALUATION OF SKIN ERYTHEMA INDUCED BY TOPICAL APPLICATION OF THE FORMULATIONS The skin erythema induced by topical application of three different concentrations (% w/w) of glycolic, mandelic, and grape acids was evaluated by refl ectance spectrophotometric mea- sures. Spectral data were recorded for each site treated with the formulations over a mon- itoring period of 50 hours. The AUC values calculated over the monitoring time are shown in Figure 2, and the trends in Δ.E.I. versus time are reported in Figure 3 (curves a, b, and c) at 10%, 30%, and 50% w/w. A signifi cant increase in skin erythema was observed from the lower to the higher concentrations for GLY, MAN, and GA. Further- more, at all three concentrations, GLY showed higher values of AUC than MAN and GA ( p 0.05). No statistical differences were found between MAN and GA at 10% w/w of concentration ( p 0.05). However, MAN induced a signifi cantly higher AUC than GA at 30% and 50% w/w ( p 0.05). IN VIVO EVALUATION OF THE PHOTOSENSITIZING EFFECTS INDUCED BY TOPICAL APPLICATION OF THE FORMULATIONS The increase in skin sensitivity to UV light, induced by topical application of the 10% GLY, MAN, and GA formulations, was expressed as the photosensitivity percentage calculated using the erythema index values obtained by the refl ectance spectrophotometric method. Table I RT (recovery time) Values for Each Subject, Obtained in Treating Skin Sites with 10% GLY (glycolic acid), 10% MAN (mandelic acid), and 10% GA (grape acids) Formulations or without Treatment (CONTR) Subjects CONTR 10% MAN 10% GA 10% GLY 1 15 12 7 6 2 15 9 7 7 3 15 12 10 8 4 15 10 10 5 5 14 10 10 9 6 14 10 6 6 7 14 9 8 8 8 15 10 9 6 9 11 15 10 7 10 15 14 7 8 11 15 13 9 9 12 13 15 10 7 13 15 9 9 4 14 14 11 8 5 15 15 9 7 7 16 13 14 11 6 17 15 13 7 7 18 14 10 7 5 19 15 12 11 4 20 14 10 8 6 Mean 14.30 11.35 8.55 6.50 S.D 1.03 2.06 1.54 1.47 The “recovery time” value (RT) was evaluated as the time (days) required to obtained an M.I. baseline value (before dihydroxyacetone treatment) for each site.