REMOVAL OF STRATUM CORNEUM BY SCRUB CREAM 113 Evaluation. For each subject the five areas were examined and ranked visually according to degree of irritation (erythema and possible infiltration). On each subject the most irritated area was given score 1 and the least score 5. Assessments were performed in a blind manner without access to the application charts. Each site was also measured for transepidermal water loss, TEWL, with an evaporimeter (Evaporimeter Ep 1, Ser- vomed, Sweden) (8) and by estimation of dermal capillary blood flow (i.e., erythema) by laser Doppler velocimetry (9, 10). The evaluations were made in the above men- tioned order. The probe of the evaporimeter, used for TEWL measurements, contains two hygro- sensors for measurement of relative humidity at two levels, about 3 and 6 mm above the surface. The sensors are protected by a Teflon tube, allowing free diffusion of water vapor between the skin and the surrounding air. From the data obtained from these sensors, the instrument calculates the water pressure gradient and expresses the water vapor flux. The probe was held in contact with the test areas for a minimum of one minute, during which time the data was recorded on a chart strip recorder (Servogor 120 BBC). The mean value after equilibration was taken as the TEWL. Cutaneous blood flow values (CBFV) were determined by laser Doppler flowmetry (Per- iflux PF1, Perimed, Sweden). The instrument measures the microvascular blood cell flux in the skin (9, 10). The expressed signal is proportional to flow velocity and appears on the panel as units of volts. The instrument was adjusted to a band width of 4 kHz and a gain of 10. The laser probe was attached to the central part of each exposed area with commercially available probe holders. These were fixed to the skin, using double- sided adhesive tape. Values were then recorded on the chart strip recorder, and the mean recorded during a two-minute reading taken as the CBFV. SURFACE TEXTURE Subjects and treatments. Eleven female subjects (mean age 42 years, range 22-59 years) without skin diseases participated in the study. Three replicas were taken of the volar forearm skin surface. The first replica was taken prior to the treatments, the second after washing the surface with a commercial liquid soap and water for 15 seconds, blotting with soft paper, and allowing the skin to dry for about five minutes. The third replica was generated after gently rubbing the skin with a scrub cream for 20-30 seconds, rinsing with water, blotting with soft paper, and allowing the skin to dry for about five minutes. Replication procedure. The skin surface structure was transferred onto a harder material in a two-step process (skin cast). A negative replica of the skin surface was generated using a silicon-based dental impression material (Povil L-Base q- L-Catalyst, Bayer Dental). This negative replica was then used as a cast for a hard positive replica (KerrTab 2000, Sybron/Kerr, Kerr Europe 1-84018 Scafati). The topographical analysis was obtained from the hard replica. Topographical analysis. A standard stylus profilometer (Perthometer C5D), together with a precision traversing table (developed at the Department of Production Engineering, Chalmers, G/3teborg) (Figure 1), was used to measure the surface (6,7) of the hard replica. An IBM-compatible XT microcomputer was used to control the table and to collect and store the data. The analogue signals (_ 5 V) of the vertical displacements

114 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS [ Perthometer t Traversing table A/D Converter[ I Computer] I [Printer[ Figure 1. Schematic drawing of the profilometry equipment used. were digitized with a 12-bit analogue-to-digital (A/D) converter. The traces were run perpendicular to the major furrows, since previous profilometry studies have demon- strated different values of the roughness parameters depending on the directions of the scans (11-13). From each replica, 75 parallel profile traces (interval 40 !zm) were collected within an area of 6.0 X 3.0 mm. Each profile trace consisted of 1024 points, i.e., the three-dimensional reconstruction of the skin surface was built up from a total of 76,800 points. Prior to the analysis, a mean plane was fitted to the data using the least square method. Surface texture parameters. Parameters describing the texture of the surface can be grouped into two categories: amplitude parameters (R•, Rq, and Ry) and shape parameters (Rsk, RK, Rn, k•, and A•). Amplitude parameters are measures of the vertical characteristics of the surface deviations, whereas shape parameters are a combination of the horizontal and vertical characteristics of the surface. The parameters are defined according to Inter- national Standards ISO 4287/1 (published by the International Organization for Stan- dardization). ß Ra (!xm): The most widely used parameter for roughness. This parameter comprises the arithmetic mean of the departures of the profile from the mean line. ß Rq (•m): The root mean square parameter corresponding to R•. ß Ry (!xm): The maximum peak-to-valley height of the entire surface. ß Rsk (skewness): The measure of the symmetry of the profile about the mean line. The value is an indicator of the cumulative width of peaks in relation to the cumulative width of the furrows, i.e., of the asymmetry of the height distribution in relation to the middle line. Skewness is negative when the relief is mainly formed of broad peaks (plateaus) and positive when the furrows mainly have horizontal bases. This param- eter is unitless. ßRlc (kurtosis): A measure of the sharpness of the surface profile. Kurtosis = 3 for a Gaussian distribution. Kurtosis 3 indicates that the base of the distribution curve is wider than a Gaussian curve. This parameter is also unitless. ß Rn: The number of peaks per cm about the mean line. ß Aa: The arithmetic mean slope in radians of the profile. ß • (wavelength): The average measure of the spacings between local peaks and valleys, taking into account their relative amplitudes and individual spatial frequencies. Statistics. Correlation between the skin response (TEWL or CBFV) and the number of