170 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Table IV Mean Skin Physical Parameter Variation With Visual Wrinkle Score and Sites Mean physical parameter --- SD (no. in group) Wrinkle Parameter score Forehead Eye corner Cheek Mouth corner Thickness 1,2 1.53 --- 0.19 (81) 1.53 --- 0.20 (48) 1.70 + 0.20 (113) 1.62 --- 0.22 (93) 3 1.59 --- 0.23 (37) 1.50 --- 0.19 (18) 1.79 --- 0.18 (42)* 1.61 --- 0.19 (52) 4 1.57 --- 0.18 (35) 1.57 --- 0.22 (67) 1.68 --- 0.16 (10) 1.68 --- 0.24 (17) 5 1.69 --- 0.23 (18)** 1.63 --- 0.19 (36)* 1.80 --- 0.20 (3) 1.61 + 0.09 (7) Ur/Uf 1,2 0.48 --- 0.12 (81) 0.58 + 0.10 (48) 0.56 --- 0.11 (88) 0.67 --- 0.12 (93) 3 0.44 --- 0.11 (35) 0.50 --- 0.12 (18)** 0.47 --- 0.06 (42)** 0.56 + 0.09 (52)** 4 0.42 + 0.11 (35)* 0.37 -+ 0.07 (67)** 0.48 + 0.09 (17)** 0.53 + 0.09 (17)** 5 0.39 -+ 0.09 (18)** 0.34 -+ 0.07 (36)** 0.44 -+ 0.08 (6)** 0.50 -+ 0.11 (7)** Ue* 1,2 0.22 + 0.12 (81) 0.28 -+ 0.14 (48) 0.23 + 0.09 (88) 0.18 + 0.10 (93) 3 0.18 -+ 0.11 (35) 0.23 + 0.14 (18) 0.19 -+ 0.09 (42) 0.14 -+ 0.06 (52)* 4 0.19 -+ 0.09 (35) 0.16 + 0.09 (67)** 0.22 -+ 0.13 (17) 0.11 + 0.04 (17)* 5 0.20 + 0.13 (18) 0.19 -+ 0.13 (36)** 0.21 + 0.09 (6) 0.13 -+ 0.07 (7) *'** Significantly different from wrinkle score of 1 and 2 (p 0.01, p 0.05). Table V Mean Physical Parameter Variation With Wrinkle Score at Several Facial Sites Mean physical parameter + SD Wrinkle score (no. in group) Parameter Site Age? 1 and 2 3-5 Thickness Eye corner 25-35 1.575 -+ 0.252 (9) 1.530 --- 0. 195 (13) Forehead 35-45 1.554 --- 0.207 (17) 1.512 --- 0.174 (25) Cheek 40-50 1.770 --- 0. 189 (30) 1.819 --- 0. 187 (10) Mouth corner 40-50 1.681 --- 0. 150 (17) 1.606 --- 0.222 (23) Ur/Uf Eye corner 25-35 0.573 --- 0.056 (9) 0.466 --- 0.098 (13)** Forehead 35-45 0.392 --- 0.108 (17) 0.452 -+ 0.121 (25) Cheek 40-50 0.506 --- 0.078 (19) 0.491 --- 0.061 (25) Mouth corner 40-50 0.578 --- 0.090 (16) 0.577 --- 0.077 (24) Ue* Eye corner 25-35 0.318 --- 0. 186 (9) 0.186 --- 0. 113 (13)* Forehead 35-45 0.179 --- 0.078 (17) 0.184 --- 0.099 (25) Cheek 40-50 0.211 --- 0.091 (19) 0. 196 --- 0.096 (25) Mouth corner 40-50 0.136 --- 0.041 (16) 0. 154 -+ 0.063 (24) **'* Significantly different from wrinkle score of 1 and 2 (p 0.01, p 0.05). Age group with average wrinkle score of 2-3 calculated from linear regression line. significantly between the two groups. At other sites, there were no changes with other physical parameters between the two groups. DISCUSSION Previously, we measured age-related changes in the elasticity of human face skin and demonstrated that intrinsic aging and actinic aging can be monitored by assessing the relative parameter Ur/Uf and absolute parameter Ue*, respectively. In this study,

FACIAL WRINKLES 171 scoring of wrinkles was performed using photographs simultaneously with the measure- ment of these parameters, and the associations between these parameters and wrinkle formation and progression were evaluated. Some changes associated with wrinkle for- mation were found. There have been previous studies on the relationship between wrinkle formation and age. Morikawa et al. (17) investigated wrinkles in the facial skin of Japanese women using a scoring system and evaluated their developmental distribution the earliest wrinkle formation was observed at the eye corners. The investigators' results on wrinkle formation in the eye corners, forehead, cheeks, and mouth angles were very similar to ours. Corcuff et al. (7) found that crow's feet patterns in American women were formed by 35 years of age and deepened continuously with advancing age. Hayashi et al. (18) measured crow's feet wrinkles in Japanese women using parameters of depth and the ratio of wrinkle area (RWA) with a replica photographing system and image analysis, and reported that the process of wrinkle formation could be classified roughly into two stages. In the initial stage of wrinkling, RWA increased greatly in the 30s, and the second stage showed accelerated increases in wrinkle depth. Our results on wrinkle formation in the eye corners were very similar to those described above. There have been some previous histological studies of wrinkles. Kligman et al. (1) reported that no histological features distinguish the various wrinkles from surrounding skin, and concluded that wrinkling is a configuration change without specific structural alterations at the histological level. Pierad and Lapiere (3) studied facial frown lines on cadaver skin and reported that the microanatomical basis lies in the hypodermis where trabeculae of the retinacula cutis were broader and much shorter underneath the wrinkles than in the surrounding skin. Tuji et al. (2) reported both light and scanning electron microscopic findings that showed less elastotic changes in the upper dermis in areas of deep wrinkles than in that of the surrounding skin. Our measurements showed that the skin was thick in subjects showing large wrinkles in the eye corners and forehead. This suggested elastosis supports the results reported by Tuji et al. (2). On the other hand, there have been only a few studies of the physical properties of face skin. Warren et al. (20) measured the physical properties and wrinkles in subjects who were frequently exposed to the sun and in those who were not. No differences were observed in the physical properties, but wrinkle formation was significantly more marked in those subjects frequently exposed to the sun. Since the changes in physical properties due to exposure to the sun are less marked than the morphological changes in the skin surface, differences in physical properties may not have been observed. We observed that physical properties showed only a slight relationship with the extent of wrinkling, despite the marked relationship with the wrinkle formation rate. Further- more, in our study, we confirmed that physical parameters and the visual wrinkle score are correlated. This correlation was similar to that with age, but the correlation coef- ficient was higher with age than with wrinkle scores. In this study, we used the slope of the primary regression line as the wrinkle formation rate. In addition, the age at which the score became 3 was calculated. When the two values were compared, similar tendencies were observed except for a change in the order between the cheeks and mouth angles. This difference may be associated with the y intercept, i.e., the Ur/Uf value in the initial stage. Wrinkles may not form when the Ur/Uf value exceeds a certain level but form readily below this level. Comparison