SKIN PENETRATION 491 tion has little effect on penetration rates but would influence the length of time a substance would act. Thus, the local vasoconstriction produced by corticosteroids would be expected to delay their removal from the skin. Penetration has been shown to be greatly enhanced when skin is abraded, broken or inflamed. Loeffler and Thomas (24) found that 50% of Sr 89 labelled strontium chloride was absorbed through abraded rat skin in contrast to 10% absorption in intact skin sites. Livingood (25) has shown that hydrocortisone-4-C14 penetrates several times faster through patches of irritative or atopic dermatitis than through normal skin. The area of application, contact time and frequency of reapplication would be expected to have a direct effect on the extent of penetration. McKenzie and Stoughton (26) have shown experimentally that penetra- tion of corticosteroids may be increased 100-fold by occluding the site of application, and thus hydrating the stratum corneum. Vickers (27) has demonstrated that occlusion not only enhances penetration of corti- costeroids but creates a depot effect in the stratum corneum. Although the penetration of many substances has been shown to be enhanced when the stratum corneumis hydrated, Brown (28) found that• percutaneous LD so of parathion was greater when the skin was occluded than when it was left uncovered. Pretreatment of the skin with organic solvents has variable effects on permeability. Treatment with ether did not alter the penetration rate of salicylates (29) or surfactants (30) whilst the polar solvents, acetone, alcohol and hexane greatly increased the penetration of water into the skin (31). It is generally considered that the physico-chemical characteristics of a penetrant are the most important factors contributing to its penetration. Solubility, molecular size, particle size, crystalline form, volatility and polarity may all influence penetration rates. Stoughton, Clendenning and Kruze (32) using a series of nicotinic acid esters, Clendenning and Stoughton (33) using a series of phenylboronic acid compounds, and Treherne (34) using a series of non-electrolytes showed that those compounds that pene- trated best had a lipid/water partition coefficient closest to one. However, some ions in aqueous solution have been shown to penetrate skin as rapidly as other aqueous solutes (34). There appears to be some correlation between molecular size and pene- tration rate, small molecules penetrating faster than large ones, within a large range of molecular sizes. However, Iunin (35) has reported the trans- ference of colloidal sulphur across rabbit skin, and Lizgunova (36) has

49• JOURNAl. OF THE SOCIETY OF COSMETIC CHEMISTS found bacteria below the skin after dipping rodents' tails in concentrated bacterial culture. A reduction of the particle size of fluocinolone acetonide has been shown to enhance its penetration (37). Where a drug exists in more than one crystalline form the one with the highest thermodynamic activity would be expected to penetrate most rapidly, provided it is stable. Cronin and Stoughton (38) using excised human skin showed that ethyl nicotinate penetrates 37,000 times better than nicotinic acid. This was attributed either to the difference in ether/water partition coefficient (ethyl nicotinate 0.136, nicotinic acid 9.25) or to the fact that ethyl nico- tinate is 10,000 times as volatile as nicotinic acid. Certainly volatility would be expected to contribute to penetration provided the substance penetrates the skin in the vapour state. If a substance penetrates in solution then volatility •vould only contribute to the initial build up at the skin surface. Schlagel (39) reviewing topical anti-inflammatory steroids has shown that, with the exception of hydrocortisone acetate, the more polar and less mobile compounds have the lesser topical potencies. Thus triamcinolone acetonide has a topical activity one thousandfold greater (40) than the more polar triamcinolone. The vehicle may enhance the penetration of a substance by making close contact with the skin surface, by being miscible with the skin lipid film and by providing an occlusive effect which would tend to hydrate the stratum corneum. Ho•vever, it is the physico-chemical relationship of penetrant to vehicle which is probably more important. The literature on the influence of vehicles on skin penetration is confusing and sometimes contradictory, firstly because a variety of experimental animals have been used, secondly because many different methods of estimating skin pene- tration have been used, and thirdly because of a lack of awareness of possible drug vehicle interactions and of the functions of different vehicles. The following factors may be considered in the penetrant vehicle relationship:- 1. The solubility of penetrant in the vehicles or a constituent of the vehicle. 2. The rate of diffusion of penetrant within the vehicle. 3. The rate of release of penetrant from the vehicle. 4. The possible release of penetrant in solubilised form together with a constituent of the vehicle. The direction of flow in a system is from higher thermodynamic potential