Mild but effective skin cleansing—Evaluation of laureth‐23 as a primary surfactant

Skin cleansing products are among the main reasons for the development of hand eczema. Therefore, a mild but effective skin cleansing product is of particular interest, especially in the work place, where various contaminations frequently have to be removed from hands.


INTRODUCTION
For years, occupational skin diseases are on top of the annual list of suspected work-related diseases and have accounted for about 30%-40% of all occupational diseases in Germany.Within this, hand eczema is one of the most frequently diagnosed occupational dermatoses [1,2].Occupational skin diseases cause high costs for the employer due to necessary rehabilitation measures, as well as retraining measures and possibly retirement.Overall, they cause annual economic costs of 1.5 billion euros per year in Germany only [1].However, occupational skin diseases also have major impact on the employee's lives.The persons effected often have a significantly reduced quality of life and even up to a quarter of the reported suspected cases lose their jobs [2][3][4].Hairdressers, metal workers, health care workers, food industry workers, construction workers, cleaners and painters are mostly affected.In the food and healthcare business, cleansing and disinfection products are among the main reasons for the development of hand eczema [1,2].In this context, the irritant effect of skin cleansers depends mainly on two factors.Firstly, it depends on how long and intense the cleanser is used.Secondly, the irritant potential is determined through the surfactant types and combinations within the cleansing system [3,5].
Surfactants primarily consist of two different parts-a hydrophilic head and a hydrophobic tail.Due to this bipolar structure, they are able to reduce surface tension at the interface between phases and therefore, they remove dirt from the skin [5][6][7].However, also because of their structure, they are capable of interacting with proteins and lipids of the stratum corneum (SC), the outer layer of the human skin.The SC functions as a barrier against environmental pollution and pathogens and prevents the skin from excessive water loss [5][6][7][8].Nevertheless, when surfactants interact with SC proteins and lipids, this can lead to a disruption of the epidermal barrier as they denature proteins and wash out necessary lipids from the skin.This can result in irritant contact dermatitis in the long run-as a consequence of frequent hand washing [7,9,10].
In the field of professional skin cleansing, sodium laureth sulfate (SLES) is one of the most commonly used surfactants [11].This anionic surfactant shows strong protein binding due to electrostatic interactions and therefore, is strongly irritating the skin.Besides low costs, SLES also shows good cleaning performance and is often used as a primary surfactant [6,[12][13][14][15].To reduce skin irritation, SLES is mostly combined with sugar-based nonionic surfactants like alkyl polyglycosides (APGs) [11].These nonionic surfactants have, like most nonionic surfactants, weak protein interactions only, but also weak cleaning performance.This is why they are normally used as cosurfactants [6,7,12].
The nonionic surfactant laureth-23 is rarely used in professional skin cleansing [11].While it is accepted that

Conclusion:
En conclusion, laureth-23 a montré de très bonnes performances en tant que surfactant primaire.L'association de laureth-23, de cocamidopropyl bêtaine et de laureth sulfosuccinate disodique s'est notamment avérée avoir peu d'impact sur la peau tout en montrant de bonnes performances de nettoyage.laureth-23 is a mild surfactant, less is known about its cleaning performance [16].Therefore, combined with suitable co-surfactants, a mild but also effective skin cleansing product could be created with laureth-23 as a primary surfactant.Hence, in this study, the skin compatibility of laureth-23 was firstly compared with two other fatty alcohol ethoxylates (FAEO) to confirm its mild character.Afterwards, different laureth-23 surfactant combinations were tested for their skin compatibility and cleaning performance as well as compared to an already known mild but effective skin cleanser.

Test substances and reference substances
The used test substances were provided by Peter Greven Physioderm GmbH, Euskirchen, Germany.An overview of the substances and concentrations is given in Table 1.All test substances were investigated by occlusive patch testing.For the cleaning studies, only surfactant combinations 11 and 12 were chosen.

Atopic cleanser (AC)
The Atopic cleanser served as a reference for an already established skin cleanser which is particularly gentle to the skin while also showing adequate skin cleansing performance for slightly adherent hydrophilic dirt.The AC was developed in a multi-centre project from the Deutsche Gesetzliche Unfallversicherung (DGUV) [17].The project showed that the AC is better balanced in terms of skin irritation and cleaning performance than other commercially available skin cleansers.Especially the skin compatibility was very good compared to other tested cleansers [17,18].

Occlusive patch testing
Twenty-four healthy volunteers aged between 18 and 55, skin type according to Fitzpatrick I-IV, participated in this study, 18 female and 6 male.The atopy score was collected and ranged from 0 to 14 (mean 3.6); two subjects therefore had an atopic diathesis (see inclusion and exclusion criteria in Table 2).Informed consent was obtained from all participants and the study was approved by the local ethics committee (file no.4/71043.5).The flexor side of the forearm was chosen as a test region.All participants were acclimatized to the standardized room air-conditioning (room temperature of 21 ± 1°C, relative air humidity [RH] of 50 ± 5% RH) for at least 30 min to allow full adaptation of the skin to the environmental conditions before visual scoring, transepidermal water loss (TEWL) and erythema were measured (T0).Then, 60 μL of each test solution were pipetted in large Finn Chambers and applied for 24 h on both forearms.Test chambers were removed after 24 h and the bioengineering parameters were evaluated by measurements of TEWL, erythema and visual scoring 48 h after patch removal (T1).

No. Primary surfactant [%] Co-surfactant [%] Co-surfactant [%]
Transepidermal water loss was quantified using TEWAmeter TM 300 (Courage & Khazaka, Cologne, Germany) according to published guidelines for TEWL measurements [24].Erythema (a*) was measured with a chromameter CR 400 from Konica Minolta (Langenhagen, Germany).In the context of dermatological questions, as in the present study, the three-dimensional colour system developed by the Commission Internationale de l'Eclairage (CIE) was used, which can be represented as a colour space sphere with three axes and where a*-axis is relevant for the irritant effects [25].
Visual scoring (VS) ranged from 0 to 4 according to the European Society of Contact Dermatitis (ESCD) guidelines on clinical scoring of acute SLS irritant reactions [26], where erythema, oedema and scaling were considered.Visual scoring was performed by a trained dermatologist under same conditions (room, light, distance).The scale was as follows: 0 = negative/no damage; 0.5 = doubtful/very weak erythema or minute scaling; 1 = weak/weak erythema, slight oedema, slight scaling and/or slight roughness; 2 = moderate/moderate degree of erythema, oedema, scaling and/or roughness; 3 = strong/marked degree of erythema, oedema, scaling, roughness; 4 = very strong/caustic/as VS 3, with necrotic areas [27].
As TEWL, a* and VS are all interesting parameters to evaluate the irritation of the skin, we summarized the results by calculating an irritation score (IRS) to consider all parameters combinatively.Therefore, the results in each category (TEWL, a* and VS) were ranked from low to high irritant potential.Afterwards, the ranking positions (rank) were added to a sum for every test substance.
Hence, a low IRS means low irritant potential whereas a higher IRS describes higher irritant potential for the surfactant or surfactant combination.By calculating the irritation score this way, each parameter was given equal weight.

Cleaning studies
Ten healthy volunteers, skin type according to Fitzpatrick I-IV, four male and six female, aged between 18 and 55 participated in the cleaning studies.Informed consent was obtained from all participants and the study was approved by the local ethics committee.For standardized cleaning of the skin, an automated cleansing device (ACiD) developed by Sonsmann et al. [28] was used.In short, the device is composed of rotating discs covered with nitrile caps in washing chambers placed onto three defined areas of the forearm.By wiping and rotating under standardized constant pressure, a homogeneous washing result was achieved.The washing time (120 s), application pressure (150 g) and rotation speed (100 rotations per minute) were electronically controlled.The machine meets the quality criteria of DIN EN ISO 13485 for medical devices.
All participants were acclimatized to the standardized room air-conditioning (room temperature of 21 ± 1°C, relative air humidity [RH] of 50 ± 5% RH) for at least 30 min to allow full adaptation of the skin to the environmental conditions before the first measurement and before application of model dirt.The model dirt was applied according to Elsner et al. [18].Therefore, at each pole of the washing fields, 0.02 mL of the dirt was applied with a 1.0-ml syringe (Omnifix®-F Tuberkulin; Braun, Melsungen, Germany).A brass roller was rolled up and down 50 times without pressure, including the applied dirt of the opposite pole.For equal distribution of the dirt, a second clean brass roller was rolled 20 times up and down without pressure.Afterwards, a cellulose swab (Paul Hartmann, Heidenheim, Germany) was put onto the field and pressed with a press-on stamp for 10 s [18].
To control the application, chromametric luminance (L) was measured three times with a chromameter (CR 400; Konica Minolta, Langenhagen, Germany) at three defined points per application field using a measuring template to guarantee testing at the same test sites, and the coefficient of variation was calculated.A coefficient of variation of <10% was considered to be acceptable whereas higher coefficients of variation led to a rejection of the test field [18].
For the cleaning procedure, the washing chambers of the skin cleansing device were placed on the forearm.Then, 1 mL of the washing solutions were pipetted into the washing chamber.Afterwards, the washing procedure was performed as described above.The dirty washing fluid was flushed away with a special pitcher containing 130 mL of water.Through pressing a cellulose swab (Paul Hartmann, Heidenheim, Germany) onto the test field for IRS = rank (TEWL) + rank a * + rank (VS) 10 s with an 800-N press-on stamp, the remaining fluid on the test field was removed [18].
For quantification of the cleaning performance, chromametric L-values from three defined points of the washing field were measured in triplicate before the application of the model dirt.The same measurement was performed after the cleaning procedure.The mean values as well as the mean value from the application control were used for calculating the cleaning factor (CF).The CF was calculated as follows wherein L1 is the luminance before dirt application, L2 the luminance after dirt application and L3 the luminance after cleaning (Figure 1).

Statistical analysis
Statistical analyses were performed with SPSS for Windows, version 24.0 (IBM Corp., Armonk, New York, USA; https:// www.ibm.com, RRID:SCR_002865).For bioengineering parameters and the VS, the nonparametric Friedmann test was used to compare the results of groups.The Wilcoxon signed-rank testwas used afterwards to compare the differences between pairs.p-values < 0.05 were regarded as statistically significant.

Erythema
At all test sites and in all subjects, the forearms were free of appearance at the beginning of the examination following visual score according to Mehling et al. [27].This is in accordance with the inclusion criteria for participation in the examination.All test sites of all subjects received a 0 as the baseline value in the visual score.Even after irritation, a median of 0 was determined for almost all test sites.It can be concluded from this that at least in half of the participants these substances did not lead to any clinical changes on the skin.An exception was test substance 5 Inclusion and exclusion criteria for study participation in both sub studies (occlusive patch testing and cleaning studies).

Inclusion criteria Exclusion criteria
• (laureth 23 and SLES) with a median of 0.5 and the positive control SLS 0.5% with a median of 1.0 in the visual score after irritation.Therefore, also mean values are described (Figure 2).The resulting ranking positions from these mean values are shown in Table 3.
Statistical analysis for the comparison of the baseline values with the values after 48 h showed no significant increase for the atopic cleanser and for test substances 7 and 3 in the comparison of the baseline values with the values after irritation.Significant increases could be detected for all other test substances.
Regarding the development of a* value, only the positive control SLS 0.5% showed a highly significant (p < 0.001***) increase.All other test substances showed no significant changes after 48 h (Figure 3).The resulting ranking positions from these median values are shown in Table 3.

Epidermal barrier
Looking at the results for TEWL, test substances 4 (p = 0.04*) and 11 (p = 0.049*) showed a significant increase in values compared to baseline.Highly significant increases could be detected for test substances 5 and 6 compared to the baseline values (p = 0.001***).The reference substance SLS 0.5% showed a highly significant increase in TEWL values in the Wilcoxon test (p < 0.001***).All other test substances showed no significant influence on the epidermal barrier (Figure 4).The resulting ranking positions from these median values are shown in Table 3.

Irritation score (IRS)
In Table 3, the different ranking positions for TEWL, a* and VS as well as the resulting IRS are shown.As expected, the control substance SLS reached highest ranking positions in all measured parameters and therefore, the highest irritation score in this study.The lowest IRS was reached for laureth-23 as a mono surfactant.

Cleaning factor (CF)
As the used model dirt was originally developed to determine the cleaning performance of abrasive skin cleansers, residual contamination was observed in all cases.The AC showed a mean cleaning factor of 37.98% and therefore the lowest CF of the three tested substances.The two other tested surfactant mixtures both showed significant higher cleaning performance (p test substance 11 vs. AC < 0.001; p test substance 12 vs.AC = 0.004) (Figure 5).

DISCUSSION
Considering the results of the different fatty alcohol ethoxylates after occlusive patch testing, laureth-23 was the one with the lowest irritant potential.Laureth-7, with the same alkyl chain length but smaller polar head group showed significantly higher irritation scores.Ceteareth-23, with the same polar head group, but longer alkyl chain, was also more irritating than laureth-23, but not as irritating as laureth-7.This proves that both, the polar head group and the alkyl chain, have significant influence on skin compatibility of FAEO, whereas the polar head group has higher influence than the alkyl chain.Iovescu et al. already showed differences in protein surfactant interactions between different FAEOs depending on the EO degree.They found that high degrees of ethoxylation are favourable because of their meander conformation within the hydrophilic group.In consequence, less H bonds are possible between protein and surfactant.Lower EO degrees have more like a zigzag conformation and therefore, they have minimal hindering when it comes to H bonds [29].This could also explain the differences in a* and VS for laureth-23 and laureth-7 in occlusive patch testing.
Further, the surfactant combinations of laureth-23 with one co-surfactant showed different irritation scores depending on the surfactant type of the co-surfactant.The SLES combinations were more irritating than the disodium laureth sulfosuccinate combinations.This was not surprising, as in general, sulfosuccinates are recognized as being less irritating than SLES [30].The laureth-23/CABP combination was even less irritating.This was also not surprising as amphoteric surfactants like CAPB are considered to be less irritating than anionic surfactants [6,7].In general, previous studies have shown that through the combination of nonionic and anionic surfactants, micelle size is increased and the resulting critical micellar concentration (CMC) is often lower than the ideal CMC as well as the CMC of the single components.This synergistic behaviour had been explained by assuming that, within the mixed micelle, ethoxylated chains of the nonionic surfactant coil around the charged head groups of the ionic surfactant and therefore, they screen the electrostatic repulsions which favours micelle formation.In consequence, a low CMC is reached which contributes to a good skin compatibility [6,31].According to Moore et al. [32] repulsion between SLS/C12En mixed micelles and zein protein increases with greater head groups and an increasing C12En concentration.Therefore, the combination of laureth-23 with mild co-surfactants might be particularly favourable in case of skin compatibility.
In the combinations of laureth-23 with two other co-surfactants, laureth-23/amphoteric/anionic was superior to laureth-23/anionic/anionic.Especially the combination of laureth-23, CAPB and disodium laureth sulfosuccinate showed very low irritation potential.Walters et al. already showed that the addition of nonionic surfactants to an anionic/amphoteric surfactant blend is particularly favourable for the skin due to the increasing size and lower charge of the resulting micelles (Figure 6).In case of laureth-23, this phenomenon might be very significant due to the high EO degree and therefore, the high screening potential as well as   the high micellar size of the resulting mixed micelles.Also, amphoteric and anionic surfactants can synergistically influence each other depending on the pH value of the solution.Thus, the resulting micelles of nonionic/amphoteric/anionic surfactants might be more stable than the micelles of nonionic/anionic/anionic, which may also contribute to the good skin compatibility results [6,7,[33][34][35].
Besides skin compatibility, the cleaning performance of surfactants is of great interest.Normally, nonionic surfactants are rarely used as primary surfactants as they only have weak cleaning potential [12].They are typically added to anionic surfactants as co-surfactants to improve skin compatibility [6,7,11,12].Instead, anionic surfactants like SLES are used as primary surfactants due to their good cleaning performance.However, both surfactant combinations with laureth-23 as primary surfactants showed superior cleaning performance compared to AC.This was surprising, as AC contains an anionic primary surfactant.Blagojević et al. have demonstrated that adding nonionic surfactants (alcohol ethoxylate, C9-C11-7EO) to a combination of SLES/α-olefin sulfonates (AOS) improved foaming properties and induced synergism in detergency.A reduction in surface tension was also observed.Another study of them investigated the addition of nonionic surfactants (amine oxide type) to an anionic/amphoteric surfactants mixture.The addition of the nonionic surfactant in those mixtures lowered both, surface tension and CMC, as well as enhanced SLES/AOS/CAPB foaming properties.They stated that the increase of amine oxide concentration has a meaningful synergistic effect on the initial formulation properties [36,37].Therefore, synergistic effects between laureth-23 and the two cosurfactants could be possible, which may have led to the good results of the cleaning study.Further studies investigating surface tension, foaming properties as well as CMC would be necessary to better understand these results.
Comparing the two different laureth-23 surfactant combinations in the cleaning study, the laureth-23/SLES/disodium laureth sulfosuccinate combination showed a slightly higher CF than laureth-23/CAPB/disodium laureth sulfosuccinate.However, concerning the CF together with the IRS, laureth-23/CAPB/disodium laureth sulfosuccinate showed best balance and was therefore, mild to the skin while also showing good cleaning performance.
In summary, the newly developed surfactant combination showed better results than the AC in case of skin compatibility as well as cleaning performance.These findings underline the outstanding performance of the newly developed surfactant combination, especially since the AC already showed significantly better performance than other commercially available skin cleansers in the multi-centre project from the DGUV [17].These findings are of great interest, because both parameters are important for the user acceptance of employees at work.
The used standardized washing procedure offers an interesting base for the evaluation and comparison of other surfactant combinations in the future.This could, for example, be used to investigate if other surfactant combinations of the laureth-23/amphoteric/anionic type show similar effects.Also, further studies investigating the surface tension as well as the CMC of the observed mixtures could be interesting to get a better understanding of the potential synergistic effects of the tested surfactant combinations.

CONCLUSION
All in all, laureth-23 seems to be an interesting primary surfactant in terms of skin compatibility as well as cleaning performance.It was shown that the combination of laureth-23 with anionic and amphoteric surfactants is particularly favourable.Therefore, the surfactant combination of laureth-23/CAPB/disodium laureth sulfosuccinate found in this study provides an interesting base for further developments in the field of mild but effective skin cleansing.

ACKNO WLE DGE MENTS
The authors would like to dedicate this article to Dr. Meike Strunk.She was one of the main contributors to this study and passed away unexpectedly on 28 February 2020.Dr. Meike Strunk was a driven and inspiring scientist at the University of Osnabrück, with a passion for research into skin physiology and effects of protective equipment on skin barrier regeneration as a novel dimension in occupational safety and health.

CONFLICT OF INTEREST STATEMENT
Anja Dick is an employee of Peter Greven Physioderm GmbH.Hermann Josef Stolz is a member of the management board of Peter Greven GmbH & Co. KG and inventor of the European patent No. 3461472.Flora Karla Sonsmann declares no conflict of interest.
Signed informed consent • Age: 18-55 years • Light skin type Fitzpatrick I-IV • Healthy skin at the test sites • Pregnancy or breastfeeding • Hair, tattoos, scars at the test sites • Acute or manifest skin diseases • Dermatological local therapy at the test site 1 week before testing • Systemic therapy in the form of antibiotics 2 weeks before testing • Systemic antihistamine therapy, vasodilators or vasoconstrictors, for example, cortisone 3 weeks before testing • Nurturing topicals 24 h prior to testing • Known sensitization to ingredients of the test products F I G U R E 1 Illustration of the cleaning measurement fields before dirt application (L1), after dirt application (L2) and after the cleaning procedure (L3).

F I G U R E 5
Mean values of the cleaning factors (CF) of the tested surfactant combinations (AC and test substances 11 and 12).Statistical analysis: t-test for paired samples (Significant results are signed; p: *≤0.05; **≤0.005;***≤0.001).F I G U R E 6 Illustration of different (mixed) micellar structures according to Walters et al.[38].T A B L E 3 Ranking positions for visual scoring (VS), erythema (a*) and transepidermal water loss (TEWL) as well as the resulting irritation scores (IRS) ranked from low to high irritant potential.