ANTIPERSPIRANT AND DEODORANT SOAP EFFICACY 355 Table I A Comparison of Mechanical Scrub (MS) and Thran Spray Gun (TSG) Methods of Bacterial Extraction on Placebo Soap Washed Axillae Mechanical Scrub s Thran Spray Gun a log CFU/cm 2 log CFU/cm 2 Panelist Left Axilla Right Axilla Difference Left Axilla Right Axilla Difference 1 5.72 7.03 - 1.31 5.42 5.16 2 6.26 6.14 0.12 5.61 5.07 3 6.32 6.11 0.21 5.59 4.66 4 6.87 6.68 0.19 5.53 5.67 5 6.67 6.89 -- 0.22 5.98 6.25 Average - 0.20 Std. Dev. 0.64 P _0.60 0.26 0.54 0.93 --0.14 --0.27 0.26 0.49 0.30 aComparison of MS to TSG for total b__acteria extracted, as follows: ave. 23•2• (log MS,eft - log TSG•at) = (log MStight - log TSG•ight)/2 resulting in X = 0.98, std. dev. = 0.21, P 0.001. With either method when bacterial concentrations from left versus right axilla were compared, no significant differences were observed. This suggests that the random assignment of two test products represents an appropriate testing model. However, significant differences were observed between numbers of total bacteria extracted using the two procedures. The MS method removed more than five times the number of microorganisms as compared to extraction with the TSG. This difference was highly significant (p 0.001), and was apparently due to more vigorous scrubbing and hence the removal of greater numbers of microorganisms. In these experiments a significant number of subjects developed erythema as a result of the pressure applied during scrubbing (approximately 100-120 g. pressure). In subsequent bacterial extraction on one axilla of 20 panelists the physical pressure exerted during scrubbing was reduced to approximately 60-70 g. pressure, resulting in a concomitant reduction in the incidence of erythema and a reduction in the total number of bacteria extracted (Table II). Table II A Comparison of Gentle Mechanical Scrub (MS) and Thran Spray Gun (TSG) Methods of Bacterial Extraction on Placebo Soap Washed Axillae Log CFU/cm 2 Log CFU/cm 2 Panelist MS minus TSG Panelist MS minus TSG 1 0.52 11 0.08 2 - 1.38 12 - 0.45 3 0.86 13 - 0.68 4 - 1.60 14 - 0.71 5 -0.18 15 1.37 6 0.04 16 0.31 7 0.12 17 0.21 8 0.91 18 -- 0.81 9 0.59 19 --0.68 10 --0.53 20 3.34 Average 0.07 Std. Dev. 1.08 P 0.8

356 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Experimental results indicated that if very gentle mechanical scrubbing is performed nearly equivalent numbers of bacteria are extracted with the two methods. A number of experiments were conducted which involved repetitive extraction of the same axillary skin site. As illustrated in Table III, regardless of the number of extractions, the MS method removed more total bacteria than did the TSG. In addition, the first extraction with the MS method tended to remove a greater percentage of the total in situ axillary bacteria than did the TSG. While the TSG tended Table III Quantitation of Aerobic Axillary Bacterial Populations as a Function of the Number of Extractions Ave. log CFU/cm 2 Panelist # log CFU CFU Extraction # 1 2 3 4 cm 2 cm 2 % of Total Bacteria Per Extraction Mechanical Scrub • 1 6.66 7.01 6.36 5.87 6.48 3.02 x 106 79.68 2 5.98 6.44 5.85 4.52 5.70 5.01 X 105 13.22 3 5.77 5.89 5.50 3.86 5.26 1.82 X 105 4.80 4 5.40 5.78 5.00 3.64 4.96 9.12 X 104 2.41 Total 3.79 X 106 Thran Spray Gun 1 6.86 6.42 5.32 4.97 5.89 7.76 X 105 44.34 2 6.46 6.18 5.27 4.42 5.72 5.25 X 105 30.00 3 6.30 5.86 4.93 4.48 5.39 2.45 X 105 14.00 4 6.05 5.90 4.96 4.31 5.31 2.04 X 105 11.66 Total 1.75 X 106 •Each extraction consisted of a single 60 second scrub with 3 mi. of buffer. to extract fewer bacteria, this does not suggest the procedure is any less sensitive or statistically less reliable in detecting reductions in a given microbial population. As a test of reproducibility and sensitivity, both test methods were used to evaluate the germicidal efficacy of a stick antiperspirant versus a placebo stick (no antimicrobial agent). Immediately following a placebo soap wash, a placebo stick or an active containing stick antiperspirant was applied to the axillae of seven subjects. Using both bacterial extraction methods, adjacent locations in each axilla were extracted 24 hours after treatment with the antiperspirant sticks. As illustrated in Table IV, a significant reduction in bacteria was observed for axillae treated with the active containing antiperspirant 24 hours following product application. Both methods detected log reductions in CFU's. The TSG demonstrated a 1.82 log decrease in axillary bacteria or 98.49% reduction in the geometric mean bacteria per cm 2, compared to a 1.02 log reduction for the MS or 90.45% reduction in geometric mean bacteria per cm 2. Despite the fact that the MS harvested three times the number of bacteria per cm 2, both methods demonstrated log reductions in axillary bacteria as well as similar percent errors and statistical probabilities. These results suggest that either method represents a satisfactory technique for measuring in situ reductions in axillary microorganisms. In a final series of experiments the TSG was utilized to evaluate the effect of a germicide containing soap bar on the levels of in situ axillary microflora. Following a one week washout period, 20 subjects participated in four days of supervised washes with placebo soap and an antimicrobial soap containing 3,4,4'-Trichlorocarbanilide.