270 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS collected by adhesive-coated disks. It was used in the past to study xerotic changes where corneocytes were not altered by chemicals (4). Due to alterations likely induced by surfactants on corneocytes, a preliminary study was undertaken to evaluate the corre- lation between two quantitative methods of evaluation of scale collections on adhesive disks. D-Squames © (Cuderm Corporation, Dallas, TX) were placed under standardized pres- sure (110 g/cm 2) and collected from the forearms of 25 individuals with normal or "sensitive" skin. They were placed on a red calibration plate (Minolta). Measurements of the color of the samples in the L*a*b* mode were made using a reflectance color- imeter (Chroma Meter © CR200, Minolta). A dry sample of D-Squames without stratum corneum served as reference. The presence of horny layer on D-Squames alters the values of colorimetry of the samples by light scattering and reflection. Color differentials were calculated between each sample and the reference material following dE*ab = [(dL*) 2 + (da*) 2 + (db*)2] •/2. Following these measurements, the same samples were stained with basic fuchsin-toluidine blue solution (Polychrome Multiple Stain © [PMS], Delasco, Council Bluffs, IA) for one minute. Color measurements were made after placing the samples on a white calibration plate (Minolta). A dry sample of D-Squames without stratum corneum, but stained with PMS, served as reference. The values of squamometry were expressed as the color differential, dE*ab, between each sample and the reference material. This evaluation will be refered to as the basic squamometry index (SQMI). Main study. Sixteen healthy Caucasian volunteers selected after data collected in the preliminary study entered a double-blind-within-subject comparison test. The protocol was a modification of the soap chamber test (2) consisting of 12 occlusive patches (19-mm Hill Top Chambers) applied three times for 30 minutes at two-hour intervals on the volar forearms (6). Test products were five soaps (So-A, So-B, So-C, So-D, So-E), four syndets (Sy-A, Sy-B, Sy-C, Sy-D), cocamidopropyl betaine (CAPB), polysorbate 20 (Tween 20), and sodium lauryl sulfate (SLS). Each product was used as a 2%-by-weight water solution. Another site was patched with tap water alone. Applications of 0.15 ml of solutions heated at 40øC were made using Hill Top Cham- bers in the morning of the test day. Thirty minutes later the chambers were removed and the test sites thoroughly rinsed, but not scrubbed, with running tap water. After patting dry, the subjects waited 1.5 hours before fresh samples (0.15 ml) were reapplied to the original site for another 30 minutes. After removal of the chambers, the test sites were rinsed again and patted dry. The subjects waited for a further 1.5 hours, when products were reapplied a third time on the same sites, followed by the same operative principles repeated as above. Two hours after the last application of test products, D-Squames were used on each test site and stained with PMS for colorimetric evaluations, in a procedure similar to that described in the preliminary experiment. Data were expressed as color differentials, dE*ab, between the material collected at a test site and that from the water-treated site. This will be refered to as the differential squamometry index (D-SQMI). CORNEOSURFAMETRY Preliminary study. Corneosurfametry entails collection of cyanoacrylate skin surface strip-

SQUAMOMETRY AND CORNEOSURFAMETRY 271 pings (CSSS) from healthy individuals and a short contact time (two hours) with a solution of cleansing agents (5). After staining samples with PMS for three minutes and air drying, reflectance colorimetry (Chromameter ©, Minolta) is used to derive values of L* and Chroma C*. In a previous study we had reported that irritant surfactants induced both a decrease in L* and an increase in Chroma C* (5). The present evaluations were conducted to better define criteria for selecting "normal" volunteers and concentrations of test products in solution. Corneosurfametry was performed using CSSS collected from the 25 volunteers enrolled in the preliminary study dealing with squamometry. One CSSS from each volunteer was sprayed with water alone, and after staining with PMS, the difference between L* and Chroma C* was calculated to derive a colorimetric index of mildness (CIM). In fact, our experience (5) suggested that CIM is over 50 for mild surfactants and lower than 30 for harsh products. Other CSSS were used for corneosurfametry with So-A and Sy-B, both at 1:100 and 5:100 water dilutions. In addition to the usual two-hour test duration, we also kept the cleanser solutions in contact with CSSS for 8, 24 and 48 hours before PMS staining and colorimetric evaluations. Main study. Corneosurfametry was performed with the 12 products used in the main study involving squamometry. Each product, at dilutions of 1:100 and 5:100 in water, was sprayed on CSSS and kept in a moist environment for two hours. After PMS staining and colorimetric measurements, the color differential, dE*ab, was calculated between each sample and a reference CSSS sprayed with water only. This evaluation will be refered to as the corneosurfametry index (CSMI). STATISTICAL ANALYSIS Means (M), standard deviations (SD), medians (Me), and coefficients of variation (V = 102 SD M- •) were calculated. Differences between series of evaluation were evaluated by the Kolmogoroff-Smirnow test. A p-value lower than 0.05 was considered signifi- cant. Regression analysis models were applied to evaluate the relationships between parameters of squamometry and corneosurfametry. The best model, i.e., linear, loga- rithmic, or exponential, was chosen on the basis of the highest coefficient of correla- tion "r." RESULTS PRELIMINARY STUDIES A significant linear relationship (r -- . 58, p (0.01) was found between the colors of unstained D-Squames placed on a red background and PMS-stained D-Squames placed on a white background (Figure 1). The range of color variations between samples was the largest for the PMS-stained material. Hence, this way of evaluation, which is likely the most sensitive, was chosen for the main study. The overall SQMI was 23.2 -•/- 9.8 (V = 42%). In this group of volunteers, individuals with sensitive skin had always a value of SQMI higher than 25. As a criterion of selection for the main study dealing with normal skin, we only considered the 16 volunteers with SQMI lower than 25, corre- sponding to a mean of 17.7 -•/- 5.5 (V = 31%). By corneosurfametry, values of CIM with water alone were quite uniform (V = 15%)