METHOD FOR TRANSEPIDERMAL WATER LOSS 109 Application of topical vehicles can either increase or decrease the hydration of the stratum corneum depending on how the vehicle alters water activity in the barrier. Oily, occlusive materials, such as petrolatum, mineral oil, lanolin, and isopropyl my- ristate, significantly decrease the rate of water loss from the skin. These agents can be considered moisturizing agents since they are able to increase skin hydration. Humec- tants, agents that act to bind water to skin, such as glycerin and propylene glycol, can increase skin moisture by interacting with the atmospheric moisture in conditions of moderate to high humidity. When the relative humidity of air decreases, humectants on the skin, however, will extract moisture from the deeper layers. Rieger and Deem have shown that humectants increase the in vitro TEWL at conditions of low humidity (30). Humectants alone, therefore, only function as a moisturizer in proper atmospheric conditions. It is conceivable that the moisturizing emulsions can still decrease TEWL, at various humidities, even with a humectant present in the formula. Decreasing TEWL presumably increases the pool of water available to hydrate the skin. It is pos- sible to determine the influence of TEWL on moisturizer effectiveness by comparing a TEWL value on untreated skin to that value obtained from moisturizer- or occlusive agent-treated skin (30). In vitro methods for TEWL measurement can be utilized to pre-evaluate the potential efficacy of moisturizers in human skin. Present in vitro methods (1,2,4,7,8, 19,30- 33) are time-consuming to use for this purpose. Another limitation is that water loss from the skin can conflict with water loss from the moisturizer. The objective of the present research was to develop a simple and accurate method that permits rapid evaluation of TEWL of potential moisturizers by a quantitative measurement of moisture loss through an animal skin membrane. This method uses a tritiated-water tracer technique which can eliminate erroneous TEWL values that may result from the water evaporation from a moisturizer. EXPERIMENTAL PROCEDURE APPARATUS DESIGN A modification of a flow-through diffusion cell originally designed by Bronaugh (34) was utilized. The membrane was full-thickness, cartilage-stripped skin taken from the ventral ear of the male Syrian golden hamster (Harlan-Sprague Dawley, Indianapolis, IN) as described by Matias (35). Tritiated water (HTO) permeated through the mem- brane into the receptor compartment of the diffusion cell. Flux (TEWL) was deter- mined, via scintillation counting, by the measurement of HTO that adsorbed on anhy- drous calcium chloride in the receptor cell. A TEWL rate-vs-time study was conducted to determine the time to steady state or equilibrium conditions at 32øC. The prepared cells were then exposed to temperature increments within the range between 5 and 37øC, and then TEWL measured at steady state. A plot of the logarithm of TEWL vs the reciprocal of absolute temperature was used to determine the energy of activation (by slope analysis) in accordance with the Arrhenius relationship. An evaluation of TEWL was conducted for skin treated with three occlusive agents: mineral oil, castor oil, and sesame oil, and skin treated with a humectant (25% glycerin

110 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS in water) by adding the 10 IxL of the agent onto the exposed skin area (0.32 cm 2) and conducting a TEWL rate-vs-time study. A skin-conditioning study was also done. Hamster ear skin was treated in vivo with daily applications (10 Ixl/ear) of mineral oil for two weeks. Animals were then sacrificed and TEWL determined for the previously treated and untreated contralateral ear skins. APPARATUS PREPARATION Syrian golden hamsters were sacrificed by means of an intraperitoneal injection of so- dium pentobarbital (65 mg per animal). The ears were removed by incision at the base with the aid of surgical scissors. The dorsal ear skin was gently pulled away from the supporting cartilage, starting at the base and extending distally. The cartilage was gently scraped off with a scalpel from the ventral side. A 10-mm diameter dermal punch was taken from the center of the stripped skin sample. A photograph of the teflon cell used for in vitro TEWL studies is shown in Figure 1. The cell was equili- brated to constant temperature at 32øC using heated water pumped through an alu- minum holding block from a 35øC water bath or from a temperature-controlled incu- bator. The 10-mm punch biopsy of the skin was placed ventral or hairy side up on the lip or edge of the donor compartment and the top holder was screwed tightly into place. The inside section of the top was free to rotate so that the top could be secured without twisting the skin. The exposed area of the skin was approximately 0.32 cm 2. With the membrane tightly fitted in place, one side of the side-arm opening was closed with a petrolatum plug. The donor compartment of the experimental cell (section un- derneath skin) was filled from the opposite side with approximately 150 IxL of 100 IxCi/ml solution (2.2 X 108 dpm/ml) containing tritiated water (HTO). This side was also closed with a petrolatum plug. A yellow Eppendorf pipet tip with a 5.0-mm section cut off from the bottom and, subsequently, attached into a 10-mm length section of ¬" O.D., Vs" I.D., V•6" wall Tygon tubing served as the receptor compart- ment. The Tygon tubing end was attached to the screwed top portion of the diffusion cell (see Figure 1). The receptor compartment was filled with approximately 0.24 g of granular, anhydrous calcium chloride, and the top covered with Parafilm. The cell was returned to the heated aluminum holding block or the incubator, and the desiccant was removed or exchanged according to the time specified. The desiccant was placed in a 20-mL scintillation vial and dissolved in 2 ml of purified water. A 500-uL aliquot sample was then placed into a new vial, dissolved in 10 mL scintillation cocktail (Aquasol-2, New England Nuclear, Boston MA), and counted for tritium content for one minute in a Model SL400 Intertechnique scintillation counter. A known volume of HTO was dissolved in water and the counts per minute (cpm) determined by scintillation counting. This was designated as the reference standard. A background count was determined for a volume of water and subsequently subtracted from the count for the reference standard. Both the reference and the blank standard had the same concentration of dissolved calcium chloride as the sample. This was to eliminate erroneous values due to quenching. The cpm was determined for the known volume of HTO and divided by density to determine the cpm/weight. An HTO count value was determined for sample and divided by the reference value, the time, and membrane area to determine the TEWL in Ixg/cm2/hr. Six replicates were prepared.