Public health concerns

Nosocomial infections and person to person transmission of pathogenic organisms has become a problem that is costing health care facilities and patients millions of dollars every year (Kampf et al., 2004a). The increasing prevalence of bacteria that are resistant to antimicrobial agents and the inadequacy of certain antimicrobials in eliminating bacteria have resulted in a need for the development of a wider variety of more effective hand cleansers and has become a significant concern in maintaining good public health (Stickler and Thomas, 1976; Haley, 1985; Aiello, 2004). Bar soap has traditionally been one of the most commonly used means of hand sanitation, particularly since the advent of anti-microbial hand soaps. It has been demonstrated that soaps that do not contain agents that are anti-microbial in nature are seldom effective at decreasing the bacterial population present on the skin (Kampf et al., 2004a). If it can be determined that the soaps in question, Lye Soap, Dial Antibacterial (Armour-Dial, USA), Dettol (Reckitt Benckiser, United Kingdom) and Cinthol (Godrej Consumer Products Ltd., India), are effective as antimicrobials in in vitro testing it would provide a justification for in vivo testing, which is necessary to determine the actual effectiveness of these products (Voss, 1975). The tested soaps are being compared with a control soap, Irish Spring (Colgate Palmolive, USA) bar soap, to determine if there is a significant difference in antibacterial activity because it does not contain any antibacterial agents. Demonstrating the effectiveness of these soaps as inhibitors of the bacterial genera that compose the transient and resident bacteria of the skin would provide both clinicians and the general populace with more tools for combating the transmission of pathogens and decreasing the incidence of nosocomial infections and community disease (Kampf et al., 2004a). This is particularly important in developing nations where simple soaps may be the only affordable or available means of maintaining good hygiene. If these soaps are not effective at inhibiting the transient and resident flora of the skin then altering the concentration of the antibacterial agents that are incorporated in these products may be necessary (McBride, 1984).

Mode of bactericidal activity

The selected soaps were chosen based on their characteristics as antimicrobial agents. Lye Soap contains alkaline compounds which may be effective in facilitating bacterial inhibition (Madden et al., 1973). Lye soap is composed of animal lard and lye. The lye used for making lye soap is made by mixing wood ash with water and obtaining the extremely caustic liquid lye that is formed (Brown et al., 1942; Environmental Protection Agency, 1995). Lye is a strong alkali primarily composed of either sodium or potassium hydroxide which is why it is such a caustic compound (Brown et al., 1942). Lye has been used traditionally as a cleansing product is recognized as an effective cleansing agent (Brown et al., 1942; Champion et al., 1918). Lye Soap needs to be evaluated because the lye content is low enough so as to ensure that the soap's alkalinity does not result in skin irritation or toxicity (Brown et al., 1942), which may affect its antibacterial properties. In addition to the potential bactericidal effect that the lye contained in the soap may have there is also a possibility that some of the many compounds contained in the oil and fat component (lard) of the soap may have a bactericidal effect.

Dial Antibacterial, Dettol and Cinthol contain the active ingredient Trichlorocarban (TCC) which is an anilide compound (McDonnell et al., 1999; Voss, 1975). Anilide compounds, and specifically TCC, act as antibacterial agents through the disruption of cellular membranes; the membrane is disrupted through the uncoupling of proton-translocators (Hamilton and Jeacock, 1972; Hamill 1984; McDonnell et al., 1999). The concentrations of Trichlorocarban in these soaps are unknown; necessitating comparison with a control and the determination of their efficacy as antimicrobials. Trichlorocarban has been reported to be an effective antibacterial; however, its efficacy when used on the skin is questionable (McDonnell et al., 1999). Uncertainties about the antimicrobial capacity of these soaps require that they be tested, especially because of the increasing resistance of bacteria to antibacterial agents (Aiello et al., 2004, McDonnell et al., 1999).

Gram positive and gram negative susceptibility

Gram negative organisms are traditionally more difficult to eradicate and to treat in infections (Kampf et al., 2004a; McDonnell et al., 1999). Gram negative organisms are also very commonly implicated in nosocomial infections with up to 64% being caused by gram negative organisms (Kampf et al., 2004b). The lipopoly-saccharide sheath encasing gram negative bacteria is an extremely effective barrier against many antimicrobial agents and in conjunction with the outer membrane of the bacteria serves to prevent the entrance and action of hydrophobic compounds. In addition to this, the inner membrane has been thought to serve as a barrier to hydrophilic molecules giving the gram negative bacteria a very effective defense against a wide range of bactericidal agents (McDonnell et al., 1999). Another factor that may be of importance for disinfection but is unlikely to relate to skin cleansing is the increased tolerance for antiseptics through the formation of biofilms by gram negative genera (McDonnell et al., 1999). Gram positive organisms are generally much more susceptible to antiseptic treatments because the peptidoglycan composing their cell walls is able to be disrupted more easily than the polysaccharide sheath encapsulating gram negative bacteria (McDonnell et al., 1999; Kampf et al., 2004a). Trichlorocarban is especially effective against gram positive organisms and also eliminates gram negative organisms but the minimum inhibitory concentrations required for the eradication of gram negative genera is much greater than for the gram positive organisms (McDonnell et al., 1999; Voss, 1975).

Further research

This study is designed to be a preliminary justification for an in vivo experiment to determine the true efficacy of these bar soaps as topical skin antimicrobial agents. If these soaps prove to be effective they may represent an inexpensive alternative to other means of disinfection. This is beneficial, especially in the case of Lye Soap because of the difficulty that bacteria will have in developing resistance to this product because it has not been used extensively as an antibacterial and because it provides another alternative to products that bacteria may become resistant to. The results that are obtained in vitro often vary widely from the actual in vivo activity necessitating a supplementary in vivo evaluation if the in vitro test results demonstrate significant inhibitory capability (Marples et al., 1974). The null hypothesis is that there will not be a significant difference between the zone of inhibition size of the test soaps, Dial Antibacterial, Cinthol, Dettol, and Lye Soap and the Irish Spring control soap (p = 0.05). The alternative hypothesis is that the rejection of the null hypothesis is demonstrative of significant bacterial inhibition.

Media

Mueller-Hinton agar plates were used (Hamill et al., 1984; Burns et al., 2000) for testing the antimicrobial capacity of the soaps analyzed. The Mueller Hinton plates were prepared using 17g of agar (Fisher Scientific) and 22g of Mueller-Hinton Broth powder (BBL) per liter of deionized water according to the instructions in the Difco Manual (1998a) and were in 9 cm sterile plastic plates to a depth of approximately 4 mm (Deacon, 1976). For the standard plate counts nutrient agar plates (Difco Manual, 1998b) were used. All cultures were grown and the appropriate dilutions were performed using TSB media (Difco Manual, 1998c).

Bacteria

Eleven genera of bacteria were tested in this study, all of which are found in the normal resident flora of the skin and which have been implicated in nosocomial infections and are members of the transient skin flora. The organisms included for testing were obtained from Presque Isle Cultures. The Gram positive organisms were Staphylococcus aureus, Staphylococcus epidermidis, Micrococcus luteus, Streptococcus mitis, Streptococcus salivarius, Bacillus cereus, and Enterococcus faecalis. The gram negative organisms were Escherichia coli, Proteus vulgaris, Pseudomonas aeruginosa and Serratia marcescens.

Inoculum

In order to standardize the inoculum size serial dilutions were performed with E. coli, S. marcescens and S. aureus after the organisms were inoculated in TSB and allowed to grow for 18 hours at 37°C and adjusted to a specific turbidity to ensure an appropriate and consistent bacterial population. A McFarland 0.5 (0.05ml 0.048M BaCl₂ in 9.95ml 0.36N H₂SO₄) standard was used to adjust the turbidity of the broth tubes after 18 hours of growth (Kerr and McHale, 2007). Following the adjustment of turbidity with the McFarland 0.5 standard the optical density of the cultures was determined using a Bausch & Laumb Spectronic 20 spectrophotometer. Serial dilutions to 10^-8 were performed in conjunction with a standard plate count to determine the approximate number of viable cells in the culture adjusted to the McFarland 0.5 standard. An optical density between 0.10 and 0.15 at 540nm in a culture (corresponding to a turbidity McFarland 0.5 standard) contains approximately 1×10^-7 CFU, which is desirable for performing antimicrobial susceptibility tests. When performing spread plating 0.1ml of culture was plated using a glass spreader after the culture was adjusted to a McFarland 0.5 standard and verified spectrophotometrically to have an optical density between 0.10 and 0.15 at a 540nm wavelength.

Soaps and concentrations

The five soaps; Lye Soap, Dial Antibacterial, Dettol, Cinthol and the non-antibacterial Irish Spring control were tested for antimicrobial capacity by agar well diffusion (Deacon, 1976; Rojas et al., 2006) on Mueller-Hinton agar (Difco labs) of approximately 4mm in depth. Four concentrations of each soap were prepared in sterile water; 0.1g/ml, 0.01g/ml, 0.001g/ml and 0.0001g/ml. The solutions were prepared by serial dilution from a 0.1g/ml stock concentration. The 0.1g/ml solution was prepared by addition of 10.0 g of soap in 100ml of warm, sterile, de-ionized water.

Agar well diffusion

After inoculation and spread plating was performed, each plate had 5 wells 9mm in diameter (Deacon, 1976) punched using a sterile cork hole borer. To four wells on each plate 0.1ml of soap solution was added. The fifth well was used as a control and had 0.1ml of sterile saline added. The soap solutions were warmed in a 55°C incubator and thoroughly mixed before addition to the wells to ensure that a uniform concentration was obtained. If necessary the soap solutions were warmed in a boiling water bath to facilitate obtaining a uniform concentration. The plates were incubated at 37°C for 24 hours after which the zones of inhibition were determined. The zones of inhibition were measured to the nearest 0.5mm (Deacon, 1976).

Replicates

The testing was conducted three times with one replicate in each test group. The same soap solutions were stored and used in each test set and replicate to ensure that a large enough sample size was obtained for analysis. For every replicate fresh bacterial sample were prepared for inoculation. The McFarland Standard used to adjust the CFU count in the inoculum was prepared no more than half an hour prior to standardization.

Determination of pH

The pH of each soap solution at each concentration was performed twice using a Beckman Expendomatic IV pH meter. Each soap solution was thoroughly mixed (with warming if necessary) to ensure a uniform soap concentration and pH level. The pH meter was calibrated before use and after temperature adjustment for the warmed solutions, (55°C) using buffers of pH 3, 5, 7, 9 and 11.

Statistical analysis

The statistical analysis of the data was performed using Minitab® 14 Statistical Software. The software was used to evaluate the data and to produce box plots to assess normality and determine whether parametric or non-parametric analysis should be used (LeBlanc, 2006). The Kruskal Wallis test was selected for a comparison of medians because the data were significantly skewed and the data sets contained outliers (LeBlanc, 2006).

Bacterial inhibition

The inhibition of bacteria was most pronounced with the highest soap concentration. The diffusion of the soap solutions into the media was clearly visible as a ring that formed around the well into which the solution had been pipetted (Figure 1). Clear zones of inhibition were formed surrounding the wells on plates cultured with the organisms that were susceptible to the antibacterials contained in the soaps (Figure 1). Lye Soap demonstrated infrequent and random inhibition. On some of the plates the growth of bacteria surrounding the wells containing Lye Soap had reduced, but not completely inhibited, growth.

Comparison of soaps

The inhibitory effect of the soaps was determined utilizing the non-parametric Kruskal Wallis test because of the skew of the data and the presence of outliers in the sample groups, as demonstrated in Figure 2. Analysis with Kruskal Wallis demonstrated that the inhibitory effect of Lye Soap was significantly different from Dial Antibacterial, Cinthol and Dettol at concentrations of 0.1g/ml (p<0.001), 0.01g/ml (p<0.001) and 0.001g/ml (p<0.001), respectively (Figure 2). Dial Antibacterial, Cinthol and Dettol soaps, when analyzed, were not be significantly different from each other because the p-value was too high to discount the null hypothesis. At concentrations of 0.1g/ml, 0.01g/ml and 0.001g/ml the Kruskal Wallis values were p = 0.661, p = 0.513 and p = 0.948 demonstrating that these soaps did not have significantly different inhibitory activity. Although soap concentrations of 0.0001g/ml were prepared and tested, they demonstrated no inhibitory effect on bacterial growth and thus were not subjected to statistical analysis.

Figure 1.

Plate demonstrating zones of inhibition of Staphylococcus aureus for a 0.1 g/ml soap Concentration. Well A is cinthol, B is dettol, C is Dial antibacterial, D is lye soap and E is a saline control.

It is also shown in Figure 2 that as the soap concentration decreased so did the diameter of the zones of inhibition. As previously stated, the 0.0001g/ml soap concentrations were not effective for inhibitory purposes suggesting that this concentration is below the minimum inhibitory concentration for the antibacterial compounds contained in each of the soaps included in the study.

The inhibitory effect of Lye Soap when compared with that of the Irish Spring control was not determined to be different because every instance of inhibition in Lye Soap was considered to be an outlier point (Figure 3). The instances of bactericidal activity in Lye Soap were limited to Gram positive bacteria.

Gram positive vs. gram negative genera

The soaps that were tested were compared for efficacy against gram positive versus gram negative organisms (Figure 4) in order to determine the relative susceptibility of each category to Lye Soap and to the unknown TCC concentration in each antibacterial soap. The Gram positive organisms were S. aureus, S. epidermidis, M. luteus, S. mitis, S. salivarius, B. cereus, and E. faecalis. The gram negative organisms are E. coli, P. vulgaris, P. aeruginosa and S. marcescens. This test was performed after observing that most of the gram negative organisms were not inhibited by any of the soaps at any of the tested concentrations. The data was analyzed using the Kruskal Wallis test and indicated definitively that there was a significant difference (p<0.001) between the inhibition of gram positive and gram negative organisms at each of the effective concentrations of the soaps that were tested (0.1g/ml, 0.01g/ml and 0.001g/ml). As anticipated the overall inhibition of bacteria decreased with decreasing concentration of soap; the trend of soaps being more effective against gram positive organisms is also clearly evident from the data (Figure 4). It was demonstrated that there was a significant difference in inhibition between gram positive and gram negative organisms at concentrations of 0.1 g/ml (p<0.001), 0.01g/ml (p<0.001) and 0.001g/ml (p<0.001).

Figure 2.

Zone of inhibition comparison of tested soaps for all test concentrations by Kruskal Wallis analysis. The box plots of the zones of inhibition from each soap with suspected antibacterial capacity are ordered from the strongest concentration (0.1g/ml) to weakest concentration (0.001g/ml) that showed antibacterial activity.

Figure 3.

Zone of inhibition comparison of lye soap and the Irish Spring control by Kruskal Wallis analysis. The diamond shapes indicate outlier data points.

Figure 4.

Comparison of zone of inhibition in Gram positive and Gram negative organisms by Kruskal Wallis analysis. The diamond shapes indicate outlier data points.

pH

Comparisons of the pH levels of each type and concentration of soap were performed graphically because determining whether the differences in pH were significant statistically does not necessarily correspond to significance with regard to inhibitory effect. The pH values were all very similar for the soaps, and in only one instance was the pH of Lye Soap greater than that of Irish Spring (at 0.1g/ml concentration) indicating that any inhibitory effect of Lye Soap would be due to a mode of action other than antagonistic pH levels, although it is possible that pH does play a role in its minimal inhibitory capacity (statistically insignificant). Irish Spring soap demonstrated no inhibition whatsoever, matching the control saline wells, and because it had very similar pH levels to those of the other soaps at each concentration, as is shown in Figure 5, it can be concluded that pH is not likely to be, in itself, a significant factor in bacterial inhibition by bar soaps.