Dermatologic, Cosmeceutic, and Cosmetic Development: Therapeutic and Novel Approaches. Walters KA, Roberts MS, eds., New York: Informa Healthcare, 2007 Dec; :385-400
The disposition of volatile products following application to the skin plays a role in both their safety and efficacy. The absorption rate determines systemic levels and skin concentrations, both of which must be factored into risk assessments. The evaporation rate determines efficacy for certain products, for example, a fine fragrance or an insect repellant. These two factors are interdependent as both depend on and, in turn, affect the residual concentration of the compound in or on the skin. The problem of predicting skin disposition of volatiles has been of interest in our laboratory for several years. We have investigated two general approaches to this problem: a well-stirred compartment or pharmacokinetic model (1 - 6) and a diffusion model (7,8). The former is simpler and has merit for interpolation within a closely related set of compounds and exposure conditions. Its application to risk assessment for perfume raw materials (or fragrance ingredients) has been discussed (6). The pharmacokinetic approach will not be further considered here; the focus of this chapter is how to make predictions using the diffusion model. A suitable starting point for a skin diffusion model for volatile chemicals is shown in Figure 1. This is a one-dimensional model composed of four layers: a vehicle layer or donor solution, stratum corneum, viable epidermis, and dermis. The upper two layers correspond to the diffusion model described in Ref. 7. If only systemic absorption estimates are required and the compound is not highly lipophilic (9), then the solution of the diffusion equation in these two layers with sink conditions at the base of the stratum corneum provides an adequate description of the problem. Analytical solutions to this problem for a number of exposure conditions may be found (7,10,11). If, on the other hand, skin and underlying tissue concentrations are of interest, it is necessary to include an explicit representation of the skin layers and (perhaps) the underlying fat and muscle. We will not consider the subdermal layers here; for a useful discussion, see Ref. 12. A distributed model for partitioning, diffusion, and clearance in the dermis has recently been described by workers from our laboratories (13,14). This model forms the basis for the viable skin model presented here. It is noteworthy that less is known about the transport properties of viable epidermis than dermis. Our working approach, discussed later, is to treat viable epidermis as unperfused dermis. Because the diffusive resistance of viable epidermis appears to be low, this assumption has a minimal impact on systemic absorption estimates. However, it does affect the tissue concentration calculation in that layer. For problems such as allergic contact dermatitis, where the putative site of action is the Langerhans cell surface in the mid-epidermis, a physiologically accurate representation of the epidermis is desirable. The cellular nature of this tissue, in contrast to the largely acellular (but fibrous) dermis, in all probability imparts to it different selectivity for transport of chemical permeants. A study of the schematic diagram shown in Figure 1 gives rise to three important questions: (1) Is it reasonable that a slab model with no internal microstructure can accurately represent skin transport? (2) How can appropriate transport parameters for the slab model be chosen prospectively? (3) How can the calculation be implemented? Each of these questions is addressed in the following sections. An example calculation for the fragrance ingredient benzyl alcohol is then presented showing the power of the technique but also revealing some of its limitations. Complexity is incurred because many small semipolar molecules like benzyl alcohol affect skin permeability, presumably by interacting with stratum corneum lipids. A general method of predicting these interactions is not yet known.