362 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS METHODS Ten percent solutions of A1C13 containing the surfactants were occlusively applied via Duhring chambers for 3 hr to two groups of six subjects each. Sweat suppression was estimated 24 hr later. The detergents were: 1) anionic--2.5% sodium lauryl sulfate (Fisher Scientific Co., Chemical Manufacturing Division, Fairlawn, N.J.) 2) cationic- 5% cetyl pyridinium chloride (ACETO Chemical Co., Flushing, N.Y.), 5% Hyamine 2389 (Rohm & Haas Co., Phila., Penn.), 5% benzalkonium chloride (ICN Pharmaceuti- cals, Inc., Plainview, N.Y.), 10% cetyltrimethylammonium bromide (supplied courtesy of Shulton, Co., Clifton, N.J.), and 10% tetramethylammonium chloride (supplied, courtesy of Shulton Co., Clifton, N.J.) 3) nonionics--10% Triton X-100 (Rohm & Hass Co., Phila., Penn.), 10% Tween-80 (Fisher Scientific Co., Chemical Manufacturing Div., Fairlawn, NJ.) and 10% Pluronic L-35 (BASF Wyandotte Corp., Wyandotte, Mich.). The plain detergent solutions and 10% aluminum chloride comprised the controls. RESULTS An average suppression of about 50% was produced by 10% A1C13 alone. SLS enhanced the anhidrosis by about 25%, accompanied by a mild inflammatory reaction. Among the cationic surfactants, only cetyltrimethylammonium bromide and tetramethylam- monium chloride exhibited appreciable augmentation, 20% and 15% respectively, again with visible irritation. The nonionic detergents resulted in only slight improvements, ranging from 15% for Triton X-100 to 7.5% for Pluronic L-35. These latter increases are not statistically different from the increases of the controls. Used alone, the detergents had no antiperspirant activity with the exception of SLS, which produced a 34% inhibition accompanied by irritation. COMMENT Ionic surfactants are chemically reactive substances which are potentially quite irritating to the skin. We deliberately selected sub-irritating concentrations, but even so, mild erythema and scaling often occurred at control sites. The irritant reaction was worsened in the presence of aluminum chloride, itself a potential irritant. Past studies have shown that damaging the skin in various ways, including chemical insults, can lead to miliaria. Injury to the acrosyringium underlies anhidrosis produced in different ways. Accordingly, the enhancement of aluminum anhidrosis by anionic and cationic surfactants is probably a non-specific damaging effect unrelated to surface tension or wetting properties of the solutions. The nonionic non-irritating surfactants were mostly without effect. STUDY 3. EFFECT OF PRE-TREATMENTS It is controversial whether removal of the surface lipids aids antiperspirant activity. Brun (9) and Schmid et al. (11) obtained a greater degree of sweat inhibition on sites delipidized by swabbing with ether-ethanol. Kligman (12) denied any enhancement of anhidrosis by lipid removal. We sought to improve effectiveness by a variety of pretreatment maneuvers even though certain of these are quite impractical.

ANTIPERSPIRANT ACTION OF ALUMINUM SALTS 363 METHODS Ten percent aqueous A1C13.6H20 was applied via chambers for 3 hr to the volar forearms of six subjects. Anhidrosis was estimated 24 hr later. The pretreatment procedures were as follows: Defatting.' 2 ml of a solvent was pipetted into a 20ram glass cup held tightly against the skin and the surface vigorously scrubbed for one minute. The fluid was then aspirated and discarded. The solvents were ethanol, acetone, ethyl ether, hexane and physiologic saline control. Hydration.' Plastic cups with a widened rim were glued onto the skin and further secured by adhesive tape. The cups were 25ram in diameter and received 5ml of deionized water through a 5mm aperture with a removable stopper. The water-filled cups remained on the skin for 1 to 3 days. Stripping.' One-inch wide Scotch tape was pressed against the skin and quickly peeled off, four times in succession. This procedure removes the outermost 4 to 5 cell layers of the stratum corneum. RESULTS A tendency toward enhancement was observed with all the procedures in no instance, however, was the effect so great as to achieve statistical significance. The solvents increased the antiperspirant activity by 5% to 10%. The scrubbing procedure itself was slightly promoting. Stripping and hydration increased the anhidrosis by about 10%. COMMENT By delipidization we did not obtain the enhancement reported by Brun (9) and Schmid et al. (11). They removed the surface lipids prior to each of several applications of aluminum chloride, a procedure very different from our 3-hour exposure. In Schmid et al.'s studies the pre-treated sites became irritated, suggesting a toxic effect. We, too, saw mild irritation on some persons, but despite this the promoting effect was slight. Removing a few cell layers of the stratum comeurn may have facilitated entrance of AICl 3 into the orifices, but the increased antiperspirant effect was marginal. One of the most familiar ways of increasing the penetration of drugs and chemicals into the skin is by hydration. This swells the horny layer and reduces diffusional resistance. In the case of a strong electrolyte, this mechanism cannot be invoked. A1Cl 3 does not cross the horny layer barrier. The slight enhancement obtained by hydration might be due to filling the ducts with water which provides the medium for diffusion into the acrosyringium. Maneuvers which cleanse or delipidize the skin surface prior to application of antiperspirants are not likely to produce perceivable benefits in increased activity. Vigorous pre-treatments would certainly be enhancing but only at the cost of irritation, an unacceptable trade-off. STUDY 4. EFFECT OF MASSAGE The effect of massage on the penetration of topically administered drugs is a neglected area of investigation. In the case of antiperspirants we and others have found that